# Kelas otomatis

Dalam banyak kasus, arsitektur yang akan digunakan dapat disimpulkan dari nama atau path model pretrained yang diberikan ke metode `from_pretrained()`. Instansiasi [AutoConfig](/docs/transformers/main/id/model_doc/auto#transformers.AutoConfig), [AutoModel](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel), atau [AutoTokenizer](/docs/transformers/main/id/model_doc/auto#transformers.AutoTokenizer) secara otomatis membuat kelas arsitektur yang sesuai. Sebagai contoh:

```python
model = AutoModel.from_pretrained("google-bert/bert-base-cased", device_map="auto")
```

Ini membuat instance model dari [BertModel](/docs/transformers/main/id/model_doc/bert#transformers.BertModel).

Setiap tugas memiliki kelas `AutoModel` yang sesuai.

## Memperluas Kelas Otomatis

Setiap kelas auto menyediakan metode yang memungkinkan kelas tersebut diperluas dengan kelas kustom. Misalnya, jika Anda mendefinisikan kelas model kustom `NewModel`, pastikan Anda juga mendefinisikan `NewModelConfig` yang sesuai. Anda kemudian dapat mendaftarkannya dengan kelas auto seperti yang ditunjukkan di bawah ini:

```python
from transformers import AutoConfig, AutoModel

AutoConfig.register("new-model", NewModelConfig)
AutoModel.register(NewModelConfig, NewModel)
```

Anda kemudian dapat menggunakan kelas auto seperti biasa.

Jika `NewModelConfig` Anda adalah subkelas dari [PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), pastikan atribut `model_type` diatur ke kunci yang sama yang Anda gunakan saat mendaftarkan config (di sini `"new-model"`).

Demikian pula, jika `NewModel` Anda adalah subkelas dari [PreTrainedModel](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel), pastikan atribut `config_class` diatur ke kelas yang sama yang Anda gunakan saat mendaftarkan model (di sini `NewModelConfig`).

## AutoConfig[[transformers.AutoConfig]]

#### transformers.AutoConfig[[transformers.AutoConfig]]

```python
transformers.AutoConfig()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/configuration_auto.py#L286)

Ini adalah kelas konfigurasi generik yang akan diinstansiasi sebagai salah satu kelas konfigurasi dari pustaka ini saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoConfig.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoConfig.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path: str | os.PathLike[str], **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/configuration_auto.py#L309)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa salah satu dari: - String, yaitu *model id* dari konfigurasi model pretrained yang dihosting di dalam model repo di huggingface.co. - Path ke *directory* yang berisi file konfigurasi yang disimpan menggunakan metode [save_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.save_pretrained), atau metode [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`. - Path ke file konfigurasi JSON *file* yang disimpan, misalnya `./my_model_directory/configuration.json`.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, defaultnya `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi serta menimpa versi yang di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

return_unused_kwargs (`bool`, *opsional*, defaultnya `False`) : Jika `False`, fungsi ini hanya mengembalikan objek konfigurasi akhir. Jika `True`, fungsi ini mengembalikan `Tuple(config, unused_kwargs)` di mana *unused_kwargs* adalah kamus yang terdiri dari pasangan kunci/nilai yang kuncinya bukan merupakan atribut konfigurasi: yaitu, bagian dari `kwargs` yang tidak digunakan untuk memperbarui `config` dan diabaikan.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

kwargs(argument kata kunci tambahan, *opsional*) : Nilai dalam kwargs dari kunci mana pun yang merupakan atribut konfigurasi akan digunakan untuk menimpa nilai yang dimuat. Perilaku terkait pasangan kunci/nilai yang kuncinya *bukan* atribut konfigurasi dikendalikan oleh parameter kata kunci `return_unused_kwargs`.

Instansiasi salah satu kelas konfigurasi dari pustaka ini dari konfigurasi model pretrained.

Kelas konfigurasi yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config yang dimuat, atau, jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **EvollaModel** -- [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) (model EvollaConfig)
- **afmoe** -- [AfmoeConfig](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeConfig) (model AfmoeConfig)
- **aimv2** -- [Aimv2Config](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2Config) (model Aimv2Config)
- **aimv2_text_model** -- [Aimv2TextConfig](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2TextConfig) (model Aimv2TextConfig)
- **aimv2_vision_model** -- [Aimv2VisionConfig](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2VisionConfig) (model Aimv2VisionConfig)
- **albert** -- [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig) (model AlbertConfig)
- **align** -- [AlignConfig](/docs/transformers/main/id/model_doc/align#transformers.AlignConfig) (model AlignConfig)
- **align_text_model** -- [AlignTextConfig](/docs/transformers/main/id/model_doc/align#transformers.AlignTextConfig) (model AlignTextConfig)
- **align_vision_model** -- [AlignVisionConfig](/docs/transformers/main/id/model_doc/align#transformers.AlignVisionConfig) (model AlignVisionConfig)
- **altclip** -- [AltCLIPConfig](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPConfig) (model AltCLIPConfig)
- **altclip_text_model** -- [AltCLIPTextConfig](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPTextConfig) (model AltCLIPTextConfig)
- **altclip_vision_model** -- [AltCLIPVisionConfig](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPVisionConfig) (model AltCLIPVisionConfig)
- **apertus** -- [ApertusConfig](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusConfig) (model ApertusConfig)
- **arcee** -- [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig) (model ArceeConfig)
- **aria** -- [AriaConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaConfig) (model AriaConfig)
- **aria_text** -- [AriaTextConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextConfig) (model AriaTextConfig)
- **audio-spectrogram-transformer** -- [ASTConfig](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTConfig) (model ASTConfig)
- **audioflamingo3** -- [AudioFlamingo3Config](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Config) (model AudioFlamingo3Config)
- **audioflamingo3_encoder** -- [AudioFlamingo3EncoderConfig](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3EncoderConfig) (model AudioFlamingo3EncoderConfig)
- **autoformer** -- [AutoformerConfig](/docs/transformers/main/id/model_doc/autoformer#transformers.AutoformerConfig) (model AutoformerConfig)
- **axk1** -- [AXK1Config](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Config) (model AXK1Config)
- **axk2** -- [AXK2Config](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Config) (model AXK2Config)
- **aya_vision** -- [AyaVisionConfig](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionConfig) (model AyaVisionConfig)
- **bailing_hybrid** -- [BailingHybridConfig](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridConfig) (model BailingHybridConfig)
- **bamba** -- [BambaConfig](/docs/transformers/main/id/model_doc/bamba#transformers.BambaConfig) (model BambaConfig)
- **bark** -- [BarkConfig](/docs/transformers/main/id/model_doc/bark#transformers.BarkConfig) (model BarkConfig)
- **bart** -- [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig) (model BartConfig)
- **beam** -- [BeamConfig](/docs/transformers/main/id/model_doc/beam#transformers.BeamConfig) (model BeamConfig)
- **beit** -- [BeitConfig](/docs/transformers/main/id/model_doc/beit#transformers.BeitConfig) (model BeitConfig)
- **bert** -- [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) (model BertConfig)
- **bert-generation** -- [BertGenerationConfig](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationConfig) (model BertGenerationConfig)
- **big_bird** -- [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) (model BigBirdConfig)
- **bigbird_pegasus** -- [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig) (model BigBirdPegasusConfig)
- **biogpt** -- [BioGptConfig](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptConfig) (model BioGptConfig)
- **bit** -- [BitConfig](/docs/transformers/main/id/model_doc/bit#transformers.BitConfig) (model BitConfig)
- **bitnet** -- [BitNetConfig](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetConfig) (model BitNetConfig)
- **blenderbot** -- [BlenderbotConfig](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotConfig) (model BlenderbotConfig)
- **blenderbot-small** -- [BlenderbotSmallConfig](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallConfig) (model BlenderbotSmallConfig)
- **blip** -- [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig) (model BlipConfig)
- **blip-2** -- [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config) (model Blip2Config)
- **blip_2_qformer** -- [Blip2QFormerConfig](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2QFormerConfig) (model Blip2QFormerConfig)
- **blip_2_vision_model** -- [Blip2VisionConfig](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2VisionConfig) (model Blip2VisionConfig)
- **blip_text_model** -- [BlipTextConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipTextConfig) (model BlipTextConfig)
- **blip_vision_model** -- [BlipVisionConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipVisionConfig) (model BlipVisionConfig)
- **bloom** -- [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) (model BloomConfig)
- **blt** -- [BltConfig](/docs/transformers/main/id/model_doc/blt#transformers.BltConfig) (model BltConfig)
- **blt_global_transformer** -- `BltGlobalTransformerConfig` (model BltGlobalTransformerConfig)
- **blt_local_decoder** -- `BltLocalDecoderConfig` (model BltLocalDecoderConfig)
- **blt_local_encoder** -- `BltLocalEncoderConfig` (model BltLocalEncoderConfig)
- **blt_patcher** -- `BltPatcherConfig` (model BltPatcherConfig)
- **bridgetower** -- [BridgeTowerConfig](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerConfig) (model BridgeTowerConfig)
- **bridgetower_text_model** -- [BridgeTowerTextConfig](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerTextConfig) (model BridgeTowerTextConfig)
- **bridgetower_vision_model** -- [BridgeTowerVisionConfig](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerVisionConfig) (model BridgeTowerVisionConfig)
- **bros** -- [BrosConfig](/docs/transformers/main/id/model_doc/bros#transformers.BrosConfig) (model BrosConfig)
- **camembert** -- [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) (model CamembertConfig)
- **canary** -- [CanaryConfig](/docs/transformers/main/id/model_doc/canary#transformers.CanaryConfig) (model CanaryConfig)
- **canary_decoder** -- [CanaryDecoderConfig](/docs/transformers/main/id/model_doc/canary#transformers.CanaryDecoderConfig) (model CanaryDecoderConfig)
- **canine** -- [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig) (model CanineConfig)
- **chameleon** -- [ChameleonConfig](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonConfig) (model ChameleonConfig)
- **chameleon_vqgan** -- [ChameleonVQVAEConfig](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonVQVAEConfig) (model ChameleonVQVAEConfig)
- **chinese_clip** -- [ChineseCLIPConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPConfig) (model ChineseCLIPConfig)
- **chinese_clip_text_model** -- [ChineseCLIPTextConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPTextConfig) (model ChineseCLIPTextConfig)
- **chinese_clip_vision_model** -- [ChineseCLIPVisionConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPVisionConfig) (model ChineseCLIPVisionConfig)
- **chmv2** -- [CHMv2Config](/docs/transformers/main/id/model_doc/chmv2#transformers.CHMv2Config) (model CHMv2Config)
- **clap** -- [ClapConfig](/docs/transformers/main/id/model_doc/clap#transformers.ClapConfig) (model ClapConfig)
- **clap_audio_model** -- [ClapAudioConfig](/docs/transformers/main/id/model_doc/clap#transformers.ClapAudioConfig) (model ClapAudioConfig)
- **clap_text_model** -- [ClapTextConfig](/docs/transformers/main/id/model_doc/clap#transformers.ClapTextConfig) (model ClapTextConfig)
- **clip** -- [CLIPConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPConfig) (model CLIPConfig)
- **clip_text_model** -- [CLIPTextConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTextConfig) (model CLIPTextConfig)
- **clip_vision_model** -- [CLIPVisionConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPVisionConfig) (model CLIPVisionConfig)
- **clipseg** -- [CLIPSegConfig](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegConfig) (model CLIPSegConfig)
- **clipseg_text_model** -- [CLIPSegTextConfig](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegTextConfig) (model CLIPSegTextConfig)
- **clipseg_vision_model** -- [CLIPSegVisionConfig](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegVisionConfig) (model CLIPSegVisionConfig)
- **clvp** -- [ClvpConfig](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpConfig) (model ClvpConfig)
- **clvp_decoder** -- [ClvpDecoderConfig](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpDecoderConfig) (model ClvpDecoderConfig)
- **clvp_encoder** -- [ClvpEncoderConfig](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpEncoderConfig) (model ClvpEncoderConfig)
- **codegen** -- [CodeGenConfig](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenConfig) (model CodeGenConfig)
- **cohere** -- [CohereConfig](/docs/transformers/main/id/model_doc/cohere#transformers.CohereConfig) (model CohereConfig)
- **cohere2** -- [Cohere2Config](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2Config) (model Cohere2Config)
- **cohere2_moe** -- [Cohere2MoeConfig](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeConfig) (model Cohere2MoeConfig)
- **cohere2_vision** -- [Cohere2VisionConfig](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionConfig) (model Cohere2VisionConfig)
- **cohere_asr** -- [CohereAsrConfig](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrConfig) (model CohereAsrConfig)
- **cohere_compass** -- [CohereCompassConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassConfig) (model CohereCompassConfig)
- **cohere_compass_text** -- [CohereCompassTextConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextConfig) (model CohereCompassTextConfig)
- **cohere_compass_vision** -- [CohereCompassVisionConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassVisionConfig) (model CohereCompassVisionConfig)
- **colmodernvbert** -- [ColModernVBertConfig](/docs/transformers/main/id/model_doc/colmodernvbert#transformers.ColModernVBertConfig) (model ColModernVBertConfig)
- **colpali** -- [ColPaliConfig](/docs/transformers/main/id/model_doc/colpali#transformers.ColPaliConfig) (model ColPaliConfig)
- **colqwen2** -- [ColQwen2Config](/docs/transformers/main/id/model_doc/colqwen2#transformers.ColQwen2Config) (model ColQwen2Config)
- **conditional_detr** -- [ConditionalDetrConfig](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrConfig) (model ConditionalDetrConfig)
- **convbert** -- [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig) (model ConvBertConfig)
- **convnext** -- [ConvNextConfig](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextConfig) (model ConvNextConfig)
- **convnextv2** -- [ConvNextV2Config](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2Config) (model ConvNextV2Config)
- **cosmos3_edge** -- [Cosmos3EdgeConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeConfig) (model Cosmos3EdgeConfig)
- **cosmos3_edge_text** -- [Cosmos3EdgeTextConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeTextConfig) (model Cosmos3EdgeTextConfig)
- **cosmos3_edge_vision** -- [Cosmos3EdgeVisionConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeVisionConfig) (model Cosmos3EdgeVisionConfig)
- **cosmos3_omni** -- [Cosmos3OmniConfig](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniConfig) (model Cosmos3OmniConfig)
- **cpmant** -- [CpmAntConfig](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntConfig) (model CpmAntConfig)
- **csm** -- [CsmConfig](/docs/transformers/main/id/model_doc/csm#transformers.CsmConfig) (model CsmConfig)
- **csm_depth_decoder_model** -- [CsmDepthDecoderConfig](/docs/transformers/main/id/model_doc/csm#transformers.CsmDepthDecoderConfig) (model CsmDepthDecoderConfig)
- **ctrl** -- [CTRLConfig](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLConfig) (model CTRLConfig)
- **cvt** -- [CvtConfig](/docs/transformers/main/id/model_doc/cvt#transformers.CvtConfig) (model CvtConfig)
- **cwm** -- [CwmConfig](/docs/transformers/main/id/model_doc/cwm#transformers.CwmConfig) (model CwmConfig)
- **d_fine** -- [DFineConfig](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineConfig) (model DFineConfig)
- **dab-detr** -- [DabDetrConfig](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrConfig) (model DabDetrConfig)
- **dac** -- [DacConfig](/docs/transformers/main/id/model_doc/dac#transformers.DacConfig) (model DacConfig)
- **data2vec-audio** -- [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig) (model Data2VecAudioConfig)
- **data2vec-text** -- [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) (model Data2VecTextConfig)
- **data2vec-vision** -- [Data2VecVisionConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionConfig) (model Data2VecVisionConfig)
- **dbrx** -- [DbrxConfig](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxConfig) (model DbrxConfig)
- **deberta** -- [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config) (model DebertaV2Config)
- **decision_transformer** -- [DecisionTransformerConfig](/docs/transformers/main/id/model_doc/decision_transformer#transformers.DecisionTransformerConfig) (model DecisionTransformerConfig)
- **deepseek_ocr2** -- [DeepseekOcr2Config](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Config) (model DeepseekOcr2Config)
- **deepseek_ocr2_encoder** -- [DeepseekOcr2VisionEncoderConfig](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2VisionEncoderConfig) (model DeepseekOcr2VisionEncoderConfig)
- **deepseek_ocr2_sam_vision_model** -- [DeepseekOcr2SamVisionConfig](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2SamVisionConfig) (model DeepseekOcr2SamVisionConfig)
- **deepseek_ocr2_text** -- [DeepseekOcr2TextConfig](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2TextConfig) (model DeepseekOcr2TextConfig)
- **deepseek_ocr2_vision** -- [DeepseekOcr2VisionConfig](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2VisionConfig) (model DeepseekOcr2VisionConfig)
- **deepseek_v2** -- [DeepseekV2Config](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Config) (model DeepseekV2Config)
- **deepseek_v3** -- [DeepseekV3Config](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Config) (model DeepseekV3Config)
- **deepseek_v32** -- [DeepseekV32Config](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32Config) (model DeepseekV32Config)
- **deepseek_v4** -- [DeepseekV4Config](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4Config) (model DeepseekV4Config)
- **deepseek_vl** -- [DeepseekVLConfig](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLConfig) (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- [DeepseekVLHybridConfig](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridConfig) (model DeepseekVLHybridConfig)
- **deformable_detr** -- [DeformableDetrConfig](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrConfig) (model DeformableDetrConfig)
- **deimv2** -- [Deimv2Config](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2Config) (model Deimv2Config)
- **deit** -- [DeiTConfig](/docs/transformers/main/id/model_doc/deit#transformers.DeiTConfig) (model DeiTConfig)
- **depth_anything** -- [DepthAnythingConfig](/docs/transformers/main/id/model_doc/depth_anything#transformers.DepthAnythingConfig) (model DepthAnythingConfig)
- **depth_pro** -- [DepthProConfig](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProConfig) (model DepthProConfig)
- **detr** -- [DetrConfig](/docs/transformers/main/id/model_doc/detr#transformers.DetrConfig) (model DetrConfig)
- **dia** -- [DiaConfig](/docs/transformers/main/id/model_doc/dia#transformers.DiaConfig) (model DiaConfig)
- **dia_decoder** -- [DiaDecoderConfig](/docs/transformers/main/id/model_doc/dia#transformers.DiaDecoderConfig) (model DiaDecoderConfig)
- **dia_encoder** -- [DiaEncoderConfig](/docs/transformers/main/id/model_doc/dia#transformers.DiaEncoderConfig) (model DiaEncoderConfig)
- **diffllama** -- [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig) (model DiffLlamaConfig)
- **diffusion_gemma** -- [DiffusionGemmaConfig](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaConfig) (model DiffusionGemmaConfig)
- **diffusion_gemma_text** -- [DiffusionGemmaTextConfig](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaTextConfig) (model DiffusionGemmaTextConfig)
- **dinat** -- [DinatConfig](/docs/transformers/main/id/model_doc/dinat#transformers.DinatConfig) (model DinatConfig)
- **dinov2** -- [Dinov2Config](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2Config) (model Dinov2Config)
- **dinov2_with_registers** -- [Dinov2WithRegistersConfig](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersConfig) (model Dinov2WithRegistersConfig)
- **dinov3_convnext** -- [DINOv3ConvNextConfig](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ConvNextConfig) (model DINOv3ConvNextConfig)
- **dinov3_vit** -- [DINOv3ViTConfig](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ViTConfig) (model DINOv3ViTConfig)
- **distilbert** -- [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig) (model DistilBertConfig)
- **doge** -- [DogeConfig](/docs/transformers/main/id/model_doc/doge#transformers.DogeConfig) (model DogeConfig)
- **donut-swin** -- [DonutSwinConfig](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinConfig) (model DonutSwinConfig)
- **dots1** -- [Dots1Config](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1Config) (model Dots1Config)
- **dpr** -- [DPRConfig](/docs/transformers/main/id/model_doc/dpr#transformers.DPRConfig) (model DPRConfig)
- **dpt** -- [DPTConfig](/docs/transformers/main/id/model_doc/dpt#transformers.DPTConfig) (model DPTConfig)
- **edgetam** -- [EdgeTamConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamConfig) (model EdgeTamConfig)
- **edgetam_mask_decoder** -- [EdgeTamMaskDecoderConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamMaskDecoderConfig) (model EdgeTamMaskDecoderConfig)
- **edgetam_prompt_encoder** -- [EdgeTamPromptEncoderConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamPromptEncoderConfig) (model EdgeTamPromptEncoderConfig)
- **edgetam_video** -- [EdgeTamVideoConfig](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoConfig) (model EdgeTamVideoConfig)
- **edgetam_video_mask_decoder** -- [EdgeTamVideoMaskDecoderConfig](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoMaskDecoderConfig) (model EdgeTamVideoMaskDecoderConfig)
- **edgetam_video_prompt_encoder** -- [EdgeTamVideoPromptEncoderConfig](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoPromptEncoderConfig) (model EdgeTamVideoPromptEncoderConfig)
- **edgetam_vision_model** -- [EdgeTamVisionConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamVisionConfig) (model EdgeTamVisionConfig)
- **efficientloftr** -- [EfficientLoFTRConfig](/docs/transformers/main/id/model_doc/efficientloftr#transformers.EfficientLoFTRConfig) (model EfficientLoFTRConfig)
- **efficientnet** -- [EfficientNetConfig](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetConfig) (model EfficientNetConfig)
- **electra** -- [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) (model ElectraConfig)
- **embedding_gemma2** -- [EmbeddingGemma2Config](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2Config) (model EmbeddingGemma2Config)
- **embedding_gemma2_text** -- [EmbeddingGemma2TextConfig](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2TextConfig) (model EmbeddingGemma2TextConfig)
- **emu3** -- [Emu3Config](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Config) (model Emu3Config)
- **emu3_text_model** -- [Emu3TextConfig](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3TextConfig) (model Emu3TextConfig)
- **emu3_vqgan** -- [Emu3VQVAEConfig](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3VQVAEConfig) (model Emu3VQVAEConfig)
- **encodec** -- [EncodecConfig](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecConfig) (model EncodecConfig)
- **encoder-decoder** -- [EncoderDecoderConfig](/docs/transformers/main/id/model_doc/encoder-decoder#transformers.EncoderDecoderConfig) (model EncoderDecoderConfig)
- **eomt** -- [EomtConfig](/docs/transformers/main/id/model_doc/eomt#transformers.EomtConfig) (model EomtConfig)
- **eomt_dinov3** -- [EomtDinov3Config](/docs/transformers/main/id/model_doc/eomt_dinov3#transformers.EomtDinov3Config) (model EomtDinov3Config)
- **ernie** -- [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) (model ErnieConfig)
- **ernie4_5** -- [Ernie4_5Config](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5Config) (model Ernie4_5Config)
- **ernie4_5_moe** -- [Ernie4_5_MoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeConfig) (model Ernie4_5_MoeConfig)
- **ernie4_5_vl_moe** -- [Ernie4_5_VLMoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeConfig) (model Ernie4_5_VLMoeConfig)
- **ernie4_5_vl_moe_text** -- [Ernie4_5_VLMoeTextConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeTextConfig) (model Ernie4_5_VLMoeTextConfig)
- **ernie4_5_vl_moe_vision** -- [Ernie4_5_VLMoeVisionConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeVisionConfig) (model Ernie4_5_VLMoeVisionConfig)
- **esm** -- [EsmConfig](/docs/transformers/main/id/model_doc/esm#transformers.EsmConfig) (model EsmConfig)
- **esmc** -- [EsmcConfig](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcConfig) (model EsmcConfig)
- **esmfold2** -- [EsmFold2Config](/docs/transformers/main/id/model_doc/esmfold2#transformers.EsmFold2Config) (model EsmFold2Config)
- **eurobert** -- [EuroBertConfig](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertConfig) (model EuroBertConfig)
- **evolla** -- [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) (model EvollaConfig)
- **exaone4** -- [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) (model Exaone4Config)
- **exaone4_5** -- [Exaone4_5_Config](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Config) (model Exaone4_5_Config)
- **exaone4_5_vision** -- [Exaone4_5_VisionConfig](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_VisionConfig) (model Exaone4_5_VisionConfig)
- **exaone_moe** -- [ExaoneMoeConfig](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeConfig) (model ExaoneMoeConfig)
- **falcon** -- [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig) (model FalconConfig)
- **falcon_h1** -- [FalconH1Config](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1Config) (model FalconH1Config)
- **falcon_mamba** -- [FalconMambaConfig](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaConfig) (model FalconMambaConfig)
- **fast_vlm** -- [FastVlmConfig](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmConfig) (model FastVlmConfig)
- **fastspeech2_conformer** -- [FastSpeech2ConformerConfig](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerConfig) (model FastSpeech2ConformerConfig)
- **fastspeech2_conformer_hifigan** -- [FastSpeech2ConformerHifiGanConfig](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerHifiGanConfig) (model FastSpeech2ConformerHifiGanConfig)
- **fastspeech2_conformer_with_hifigan** -- [FastSpeech2ConformerWithHifiGanConfig](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerWithHifiGanConfig) (model FastSpeech2ConformerWithHifiGanConfig)
- **flaubert** -- [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig) (model FlaubertConfig)
- **flava** -- [FlavaConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaConfig) (model FlavaConfig)
- **flava_image_codebook** -- [FlavaImageCodebookConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaImageCodebookConfig) (model FlavaImageCodebookConfig)
- **flava_image_model** -- [FlavaImageConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaImageConfig) (model FlavaImageConfig)
- **flava_multimodal_model** -- [FlavaMultimodalConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaMultimodalConfig) (model FlavaMultimodalConfig)
- **flava_text_model** -- [FlavaTextConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaTextConfig) (model FlavaTextConfig)
- **flex_olmo** -- [FlexOlmoConfig](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoConfig) (model FlexOlmoConfig)
- **flint** -- [FlintConfig](/docs/transformers/main/id/model_doc/flint#transformers.FlintConfig) (model FlintConfig)
- **florence2** -- [Florence2Config](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Config) (model Florence2Config)
- **florence_vision** -- [Florence2VisionConfig](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2VisionConfig) (model Florence2VisionConfig)
- **fnet** -- [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig) (model FNetConfig)
- **focalnet** -- [FocalNetConfig](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetConfig) (model FocalNetConfig)
- **fsmt** -- [FSMTConfig](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTConfig) (model FSMTConfig)
- **fun_asr_nano** -- [FunAsrNanoConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoConfig) (model FunAsrNanoConfig)
- **fun_asr_nano_encoder** -- [FunAsrNanoEncoderConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoEncoderConfig) (model FunAsrNanoEncoderConfig)
- **funnel** -- [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig) (model FunnelConfig)
- **fuyu** -- [FuyuConfig](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuConfig) (model FuyuConfig)
- **gemma** -- [GemmaConfig](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaConfig) (model GemmaConfig)
- **gemma2** -- [Gemma2Config](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Config) (model Gemma2Config)
- **gemma3** -- [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) (model Gemma3Config)
- **gemma3_text** -- [Gemma3TextConfig](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextConfig) (model Gemma3TextConfig)
- **gemma3n** -- [Gemma3nConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nConfig) (model Gemma3nConfig)
- **gemma3n_audio** -- [Gemma3nAudioConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nAudioConfig) (model Gemma3nAudioConfig)
- **gemma3n_text** -- [Gemma3nTextConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nTextConfig) (model Gemma3nTextConfig)
- **gemma3n_vision** -- [Gemma3nVisionConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nVisionConfig) (model Gemma3nVisionConfig)
- **gemma4** -- [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) (model Gemma4Config)
- **gemma4_assistant** -- [Gemma4AssistantConfig](/docs/transformers/main/id/model_doc/gemma4_assistant#transformers.Gemma4AssistantConfig) (model Gemma4AssistantConfig)
- **gemma4_audio** -- [Gemma4AudioConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioConfig) (model Gemma4AudioConfig)
- **gemma4_text** -- [Gemma4TextConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4TextConfig) (model Gemma4TextConfig)
- **gemma4_unified** -- [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) (model Gemma4UnifiedConfig)
- **gemma4_unified_assistant** -- [Gemma4UnifiedAssistantConfig](/docs/transformers/main/id/model_doc/gemma4_unified_assistant#transformers.Gemma4UnifiedAssistantConfig) (model Gemma4UnifiedAssistantConfig)
- **gemma4_unified_audio** -- [Gemma4UnifiedAudioConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedAudioConfig) (model Gemma4UnifiedAudioConfig)
- **gemma4_unified_text** -- [Gemma4UnifiedTextConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedTextConfig) (model Gemma4UnifiedTextConfig)
- **gemma4_unified_vision** -- [Gemma4UnifiedVisionConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedVisionConfig) (model Gemma4UnifiedVisionConfig)
- **gemma4_vision** -- [Gemma4VisionConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4VisionConfig) (model Gemma4VisionConfig)
- **git** -- [GitConfig](/docs/transformers/main/id/model_doc/git#transformers.GitConfig) (model GitConfig)
- **git_vision_model** -- [GitVisionConfig](/docs/transformers/main/id/model_doc/git#transformers.GitVisionConfig) (model GitVisionConfig)
- **glm** -- [GlmConfig](/docs/transformers/main/id/model_doc/glm#transformers.GlmConfig) (model GlmConfig)
- **glm4** -- [Glm4Config](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Config) (model Glm4Config)
- **glm46v** -- [Glm46VConfig](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VConfig) (model Glm46VConfig)
- **glm4_moe** -- [Glm4MoeConfig](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeConfig) (model Glm4MoeConfig)
- **glm4_moe_lite** -- [Glm4MoeLiteConfig](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteConfig) (model Glm4MoeLiteConfig)
- **glm4v** -- [Glm4vConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vConfig) (model Glm4vConfig)
- **glm4v_moe** -- [Glm4vMoeConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeConfig) (model Glm4vMoeConfig)
- **glm4v_moe_text** -- [Glm4vMoeTextConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeTextConfig) (model Glm4vMoeTextConfig)
- **glm4v_moe_vision** -- [Glm4vMoeVisionConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeVisionConfig) (model Glm4vMoeVisionConfig)
- **glm4v_text** -- [Glm4vTextConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vTextConfig) (model Glm4vTextConfig)
- **glm4v_vision** -- [Glm4vVisionConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vVisionConfig) (model Glm4vVisionConfig)
- **glm5_next** -- [Glm5NextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextConfig) (model Glm5NextConfig)
- **glm5_next_text** -- [Glm5NextTextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextTextConfig) (model Glm5NextTextConfig)
- **glm5_next_vision** -- [Glm5NextVisionConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextVisionConfig) (model Glm5NextVisionConfig)
- **glm_image** -- [GlmImageConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageConfig) (model GlmImageConfig)
- **glm_image_text** -- [GlmImageTextConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageTextConfig) (model GlmImageTextConfig)
- **glm_image_vision** -- [GlmImageVisionConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVisionConfig) (model GlmImageVisionConfig)
- **glm_image_vqmodel** -- [GlmImageVQVAEConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVQVAEConfig) (model GlmImageVQVAEConfig)
- **glm_moe_dsa** -- [GlmMoeDsaConfig](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaConfig) (model GlmMoeDsaConfig)
- **glm_ocr** -- [GlmOcrConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrConfig) (model GlmOcrConfig)
- **glm_ocr_text** -- [GlmOcrTextConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrTextConfig) (model GlmOcrTextConfig)
- **glm_ocr_vision** -- [GlmOcrVisionConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrVisionConfig) (model GlmOcrVisionConfig)
- **glmasr** -- [GlmAsrConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrConfig) (model GlmAsrConfig)
- **glmasr_encoder** -- [GlmAsrEncoderConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrEncoderConfig) (model GlmAsrEncoderConfig)
- **glmga** -- [GlmgaConfig](/docs/transformers/main/id/model_doc/glmga#transformers.GlmgaConfig) (model GlmgaConfig)
- **glpn** -- [GLPNConfig](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNConfig) (model GLPNConfig)
- **got_ocr2** -- [GotOcr2Config](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Config) (model GotOcr2Config)
- **got_ocr2_vision** -- [GotOcr2VisionConfig](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2VisionConfig) (model GotOcr2VisionConfig)
- **gpt-sw3** -- [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) (model GPT2Config)
- **gpt2** -- [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig) (model GPTNeoXConfig)
- **gpt_neox_japanese** -- [GPTNeoXJapaneseConfig](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseConfig) (model GPTNeoXJapaneseConfig)
- **gpt_oss** -- [GptOssConfig](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssConfig) (model GptOssConfig)
- **gptj** -- [GPTJConfig](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJConfig) (model GPTJConfig)
- **gradium_phonon_tts** -- [GradiumPhononTTSConfig](/docs/transformers/main/id/model_doc/gradium_phonon_tts#transformers.GradiumPhononTTSConfig) (model GradiumPhononTTSConfig)
- **granite** -- [GraniteConfig](/docs/transformers/main/id/model_doc/granite#transformers.GraniteConfig) (model GraniteConfig)
- **granite4_vision** -- [Granite4VisionConfig](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionConfig) (model Granite4VisionConfig)
- **granite4_vision_text** -- [Granite4VisionTextConfig](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionTextConfig) (model Granite4VisionTextConfig)
- **granite_speech** -- [GraniteSpeechConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechConfig) (model GraniteSpeechConfig)
- **granite_speech5_ctc** -- [GraniteSpeech5CTCConfig](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5CTCConfig) (model GraniteSpeech5CTCConfig)
- **granite_speech5_encoder** -- [GraniteSpeech5EncoderConfig](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5EncoderConfig) (model GraniteSpeech5EncoderConfig)
- **granite_speech_encoder** -- [GraniteSpeechEncoderConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechEncoderConfig) (model GraniteSpeechEncoderConfig)
- **granite_speech_plus** -- [GraniteSpeechPlusConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusConfig) (model GraniteSpeechPlusConfig)
- **granite_speech_plus_encoder** -- [GraniteSpeechPlusEncoderConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusEncoderConfig) (model GraniteSpeechPlusEncoderConfig)
- **granite_swa** -- [GraniteSWAConfig](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAConfig) (model GraniteSWAConfig)
- **granitemoe** -- [GraniteMoeConfig](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeConfig) (model GraniteMoeConfig)
- **granitemoe_swa** -- [GraniteMoeSWAConfig](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAConfig) (model GraniteMoeSWAConfig)
- **granitemoehybrid** -- [GraniteMoeHybridConfig](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridConfig) (model GraniteMoeHybridConfig)
- **granitemoeshared** -- [GraniteMoeSharedConfig](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedConfig) (model GraniteMoeSharedConfig)
- **grounding-dino** -- [GroundingDinoConfig](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoConfig) (model GroundingDinoConfig)
- **groupvit** -- [GroupViTConfig](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTConfig) (model GroupViTConfig)
- **groupvit_text_model** -- [GroupViTTextConfig](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTTextConfig) (model GroupViTTextConfig)
- **groupvit_vision_model** -- [GroupViTVisionConfig](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTVisionConfig) (model GroupViTVisionConfig)
- **gte** -- [GteConfig](/docs/transformers/main/id/model_doc/gte#transformers.GteConfig) (model GteConfig)
- **helium** -- [HeliumConfig](/docs/transformers/main/id/model_doc/helium#transformers.HeliumConfig) (model HeliumConfig)
- **hgnet_v2** -- [HGNetV2Config](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2Config) (model HGNetV2Config)
- **hiera** -- [HieraConfig](/docs/transformers/main/id/model_doc/hiera#transformers.HieraConfig) (model HieraConfig)
- **higgs_audio_v2** -- [HiggsAudioV2Config](/docs/transformers/main/id/model_doc/higgs_audio_v2#transformers.HiggsAudioV2Config) (model HiggsAudioV2Config)
- **higgs_audio_v2_tokenizer** -- [HiggsAudioV2TokenizerConfig](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerConfig) (model HiggsAudioV2TokenizerConfig)
- **hrm_text** -- [HrmTextConfig](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextConfig) (model HrmTextConfig)
- **hubert** -- [HubertConfig](/docs/transformers/main/id/model_doc/hubert#transformers.HubertConfig) (model HubertConfig)
- **hunyuan_v1_dense** -- [HunYuanDenseV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Config) (model HunYuanDenseV1Config)
- **hunyuan_v1_moe** -- [HunYuanMoEV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Config) (model HunYuanMoEV1Config)
- **hunyuan_vl** -- [HunYuanVLConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLConfig) (model HunYuanVLConfig)
- **hunyuan_vl_text** -- [HunYuanVLTextConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLTextConfig) (model HunYuanVLTextConfig)
- **hunyuan_vl_vision** -- [HunYuanVLVisionConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLVisionConfig) (model HunYuanVLVisionConfig)
- **hy_v3** -- [HYV3Config](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3Config) (model HYV3Config)
- **hy_v4** -- [HYV4Config](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4Config) (model HYV4Config)
- **hyperclovax** -- [HyperCLOVAXConfig](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXConfig) (model HyperCLOVAXConfig)
- **hyperclovax_vision_v2** -- [HyperCLOVAXVisionV2Config](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Config) (model HyperCLOVAXVisionV2Config)
- **ibert** -- [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig) (model IBertConfig)
- **idefics** -- [IdeficsConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsConfig) (model IdeficsConfig)
- **idefics2** -- [Idefics2Config](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Config) (model Idefics2Config)
- **idefics2_perceiver** -- [Idefics2PerceiverConfig](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2PerceiverConfig) (model Idefics2PerceiverConfig)
- **idefics2_vision** -- [Idefics2VisionConfig](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2VisionConfig) (model Idefics2VisionConfig)
- **idefics3** -- [Idefics3Config](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Config) (model Idefics3Config)
- **idefics3_vision** -- [Idefics3VisionConfig](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3VisionConfig) (model Idefics3VisionConfig)
- **idefics_perciever** -- [IdeficsPerceiverConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsPerceiverConfig) (model IdeficsPerceiverConfig)
- **idefics_vision** -- [IdeficsVisionConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsVisionConfig) (model IdeficsVisionConfig)
- **ijepa** -- [IJepaConfig](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaConfig) (model IJepaConfig)
- **imagegpt** -- [ImageGPTConfig](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTConfig) (model ImageGPTConfig)
- **informer** -- [InformerConfig](/docs/transformers/main/id/model_doc/informer#transformers.InformerConfig) (model InformerConfig)
- **inkling_audio** -- [InklingAudioConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingAudioConfig) (model InklingAudioConfig)
- **inkling_mm_model** -- [InklingConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingConfig) (model InklingConfig)
- **inkling_text** -- [InklingTextConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingTextConfig) (model InklingTextConfig)
- **inkling_vision** -- [InklingVisionConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingVisionConfig) (model InklingVisionConfig)
- **instructblip** -- [InstructBlipConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipConfig) (model InstructBlipConfig)
- **instructblip_qformer** -- [InstructBlipQFormerConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipQFormerConfig) (model InstructBlipQFormerConfig)
- **instructblip_vision_model** -- [InstructBlipVisionConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipVisionConfig) (model InstructBlipVisionConfig)
- **instructblipvideo** -- [InstructBlipVideoConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoConfig) (model InstructBlipVideoConfig)
- **instructblipvideo_qformer** -- [InstructBlipVideoQFormerConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoQFormerConfig) (model InstructBlipVideoQFormerConfig)
- **instructblipvideo_vision_model** -- [InstructBlipVideoVisionConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoVisionConfig) (model InstructBlipVideoVisionConfig)
- **internvl** -- [InternVLConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLConfig) (model InternVLConfig)
- **internvl_vision** -- [InternVLVisionConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLVisionConfig) (model InternVLVisionConfig)
- **jais2** -- [Jais2Config](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2Config) (model Jais2Config)
- **jamba** -- [JambaConfig](/docs/transformers/main/id/model_doc/jamba#transformers.JambaConfig) (model JambaConfig)
- **janus** -- [JanusConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusConfig) (model JanusConfig)
- **janus_vision_model** -- [JanusVisionConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusVisionConfig) (model JanusVisionConfig)
- **janus_vqgan** -- [JanusVQVAEConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusVQVAEConfig) (model JanusVQVAEConfig)
- **jetmoe** -- [JetMoeConfig](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeConfig) (model JetMoeConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config) (model JinaEmbeddingsV3Config)
- **kimi_k25** -- [Kimi_K25Config](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Config) (model Kimi_K25Config)
- **kimi_k25_vision** -- [Kimi_K25VisionConfig](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25VisionConfig) (model Kimi_K25VisionConfig)
- **kimi_linear** -- [KimiLinearConfig](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearConfig) (model KimiLinearConfig)
- **kolibri1** -- [Kolibri1Config](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1Config) (model Kolibri1Config)
- **kosmos-2** -- [Kosmos2Config](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Config) (model Kosmos2Config)
- **kosmos-2.5** -- [Kosmos2_5Config](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Config) (model Kosmos2_5Config)
- **kosmos_2_5_text_model** -- [Kosmos2_5TextConfig](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5TextConfig) (model Kosmos2_5TextConfig)
- **kosmos_2_5_vision_model** -- [Kosmos2_5VisionConfig](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5VisionConfig) (model Kosmos2_5VisionConfig)
- **kosmos_2_text_model** -- [Kosmos2TextConfig](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2TextConfig) (model Kosmos2TextConfig)
- **kosmos_2_vision_model** -- [Kosmos2VisionConfig](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2VisionConfig) (model Kosmos2VisionConfig)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextConfig](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextConfig) (model KyutaiSpeechToTextConfig)
- **laguna** -- [LagunaConfig](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaConfig) (model LagunaConfig)
- **lasr_ctc** -- [LasrCTCConfig](/docs/transformers/main/id/model_doc/lasr#transformers.LasrCTCConfig) (model LasrCTCConfig)
- **lasr_encoder** -- [LasrEncoderConfig](/docs/transformers/main/id/model_doc/lasr#transformers.LasrEncoderConfig) (model LasrEncoderConfig)
- **layoutlm** -- [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config) (model LayoutLMv3Config)
- **layoutxlm** -- [LayoutXLMConfig](/docs/transformers/main/id/model_doc/layoutxlm#transformers.LayoutXLMConfig) (model LayoutXLMConfig)
- **led** -- [LEDConfig](/docs/transformers/main/id/model_doc/led#transformers.LEDConfig) (model LEDConfig)
- **levit** -- [LevitConfig](/docs/transformers/main/id/model_doc/levit#transformers.LevitConfig) (model LevitConfig)
- **lfm2** -- [Lfm2Config](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2Config) (model Lfm2Config)
- **lfm2_moe** -- [Lfm2MoeConfig](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeConfig) (model Lfm2MoeConfig)
- **lfm2_vl** -- [Lfm2VlConfig](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlConfig) (model Lfm2VlConfig)
- **lfm3** -- [Lfm3Config](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3Config) (model Lfm3Config)
- **lightglue** -- [LightGlueConfig](/docs/transformers/main/id/model_doc/lightglue#transformers.LightGlueConfig) (model LightGlueConfig)
- **lighton_ocr** -- [LightOnOcrConfig](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrConfig) (model LightOnOcrConfig)
- **lilt** -- [LiltConfig](/docs/transformers/main/id/model_doc/lilt#transformers.LiltConfig) (model LiltConfig)
- **llama** -- [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig) (model LlamaConfig)
- **llama4** -- [Llama4Config](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Config) (model Llama4Config)
- **llama4_text** -- [Llama4TextConfig](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4TextConfig) (model Llama4TextConfig)
- **llama4_vision_model** -- [Llama4VisionConfig](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4VisionConfig) (model Llama4VisionConfig)
- **llava** -- [LlavaConfig](/docs/transformers/main/id/model_doc/llava#transformers.LlavaConfig) (model LlavaConfig)
- **llava_next** -- [LlavaNextConfig](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextConfig) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoConfig](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoConfig) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionConfig](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionConfig) (model LlavaOnevisionConfig)
- **longcat_flash** -- [LongcatFlashConfig](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashConfig) (model LongcatFlashConfig)
- **longformer** -- [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig) (model LongformerConfig)
- **longt5** -- [LongT5Config](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5Config) (model LongT5Config)
- **luke** -- [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig) (model LukeConfig)
- **lw_detr** -- [LwDetrConfig](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrConfig) (model LwDetrConfig)
- **lw_detr_vit** -- [LwDetrViTConfig](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrViTConfig) (model LwDetrViTConfig)
- **lxmert** -- [LxmertConfig](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertConfig) (model LxmertConfig)
- **m2m_100** -- [M2M100Config](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Config) (model M2M100Config)
- **mamba** -- [MambaConfig](/docs/transformers/main/id/model_doc/mamba#transformers.MambaConfig) (model MambaConfig)
- **mamba2** -- [Mamba2Config](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Config) (model Mamba2Config)
- **marian** -- [MarianConfig](/docs/transformers/main/id/model_doc/marian#transformers.MarianConfig) (model MarianConfig)
- **markuplm** -- [MarkupLMConfig](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMConfig) (model MarkupLMConfig)
- **mask2former** -- [Mask2FormerConfig](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerConfig) (model Mask2FormerConfig)
- **maskformer** -- [MaskFormerConfig](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerConfig) (model MaskFormerConfig)
- **maskformer-swin** -- `MaskFormerSwinConfig` (model MaskFormerSwinConfig)
- **mbart** -- [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig) (model MBartConfig)
- **megatron-bert** -- [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) (model MegatronBertConfig)
- **mellum** -- [MellumConfig](/docs/transformers/main/id/model_doc/mellum#transformers.MellumConfig) (model MellumConfig)
- **metaclip_2** -- [MetaClip2Config](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Config) (model MetaClip2Config)
- **metaclip_2_text_model** -- [MetaClip2TextConfig](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2TextConfig) (model MetaClip2TextConfig)
- **metaclip_2_vision_model** -- [MetaClip2VisionConfig](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2VisionConfig) (model MetaClip2VisionConfig)
- **mgp-str** -- [MgpstrConfig](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrConfig) (model MgpstrConfig)
- **mimi** -- [MimiConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiConfig) (model MimiConfig)
- **mimi_decoder** -- [MimiDecoderConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiDecoderConfig) (model MimiDecoderConfig)
- **mimi_encoder** -- [MimiEncoderConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiEncoderConfig) (model MimiEncoderConfig)
- **mimo_v2_flash** -- [MiMoV2FlashConfig](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashConfig) (model MiMoV2FlashConfig)
- **minicpm3** -- [MiniCPM3Config](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Config) (model MiniCPM3Config)
- **minicpm_ocr** -- [MiniCPMOcrConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrConfig) (model MiniCPMOcrConfig)
- **minicpm_ocr_vision** -- [MiniCPMOcrVisionConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrVisionConfig) (model MiniCPMOcrVisionConfig)
- **minicpmv4_6** -- [MiniCPMV4_6Config](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Config) (model MiniCPMV4_6Config)
- **minicpmv4_6_vision** -- [MiniCPMV4_6VisionConfig](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6VisionConfig) (model MiniCPMV4_6VisionConfig)
- **minicpmv4_7** -- [MiniCPMV4_7Config](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Config) (model MiniCPMV4_7Config)
- **minicpmv4_7_vision** -- [MiniCPMV4_7VisionConfig](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7VisionConfig) (model MiniCPMV4_7VisionConfig)
- **minimax** -- [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig) (model MiniMaxConfig)
- **minimax_m2** -- [MiniMaxM2Config](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2Config) (model MiniMaxM2Config)
- **minimax_m3_vl** -- [MiniMaxM3VLConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLConfig) (model MiniMaxM3VLConfig)
- **minimax_m3_vl_text** -- [MiniMaxM3VLTextConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLTextConfig) (model MiniMaxM3VLTextConfig)
- **minimax_m3_vl_vision** -- [MiniMaxM3VLVisionConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLVisionConfig) (model MiniMaxM3VLVisionConfig)
- **ministral** -- [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig) (model MinistralConfig)
- **ministral3** -- [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config) (model Ministral3Config)
- **mistral** -- [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig) (model MistralConfig)
- **mistral3** -- [Mistral3Config](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Config) (model Mistral3Config)
- **mistral4** -- [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) (model Mistral4Config)
- **mistral_large4** -- [MistralLarge4Config](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Config) (model MistralLarge4Config)
- **mistral_large4_text** -- [MistralLarge4TextConfig](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4TextConfig) (model MistralLarge4TextConfig)
- **mixtral** -- [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig) (model MixtralConfig)
- **mlcd** -- [MLCDVisionConfig](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionConfig) (model MLCDVisionConfig)
- **mlcd_vision_model** -- [MLCDVisionConfig](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionConfig) (model MLCDVisionConfig)
- **mllama** -- [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) (model MllamaConfig)
- **mllama_text_model** -- [MllamaTextConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaTextConfig) (model MllamaTextConfig)
- **mllama_vision_model** -- [MllamaVisionConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaVisionConfig) (model MllamaVisionConfig)
- **mm-grounding-dino** -- [MMGroundingDinoConfig](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoConfig) (model MMGroundingDinoConfig)
- **mobilebert** -- [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig) (model MobileBertConfig)
- **mobilenet_v1** -- [MobileNetV1Config](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1Config) (model MobileNetV1Config)
- **mobilenet_v2** -- [MobileNetV2Config](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Config) (model MobileNetV2Config)
- **mobilevit** -- [MobileViTConfig](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTConfig) (model MobileViTConfig)
- **mobilevitv2** -- [MobileViTV2Config](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Config) (model MobileViTV2Config)
- **modernbert** -- [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig) (model ModernBertConfig)
- **modernbert-decoder** -- [ModernBertDecoderConfig](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderConfig) (model ModernBertDecoderConfig)
- **modernvbert** -- [ModernVBertConfig](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertConfig) (model ModernVBertConfig)
- **molmo2** -- [Molmo2Config](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Config) (model Molmo2Config)
- **molmo2_adapter** -- [Molmo2AdapterConfig](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2AdapterConfig) (model Molmo2AdapterConfig)
- **molmo2_text** -- [Molmo2TextConfig](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2TextConfig) (model Molmo2TextConfig)
- **molmo2_vision** -- [Molmo2VisionConfig](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2VisionConfig) (model Molmo2VisionConfig)
- **moonshine** -- [MoonshineConfig](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineConfig) (model MoonshineConfig)
- **moonshine_streaming** -- [MoonshineStreamingConfig](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingConfig) (model MoonshineStreamingConfig)
- **moonshine_streaming_encoder** -- [MoonshineStreamingEncoderConfig](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingEncoderConfig) (model MoonshineStreamingEncoderConfig)
- **moshi** -- [MoshiConfig](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiConfig) (model MoshiConfig)
- **moshi_depth** -- [MoshiDepthConfig](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiDepthConfig) (model MoshiDepthConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig) (model MossTranscribeDiarizeConfig)
- **moss_transcribe_diarize_audio** -- [MossTranscribeDiarizeAudioConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeAudioConfig) (model MossTranscribeDiarizeAudioConfig)
- **mpnet** -- [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig) (model MPNetConfig)
- **mpt** -- [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) (model MptConfig)
- **mra** -- [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig) (model MraConfig)
- **mt5** -- [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config) (model MT5Config)
- **muse_glimmer** -- [MuseGlimmerConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerConfig) (model MuseGlimmerConfig)
- **muse_glimmer_assistant** -- [MuseGlimmerAssistantConfig](/docs/transformers/main/id/model_doc/muse_glimmer_assistant#transformers.MuseGlimmerAssistantConfig) (model MuseGlimmerAssistantConfig)
- **muse_glimmer_text** -- [MuseGlimmerTextConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerTextConfig) (model MuseGlimmerTextConfig)
- **muse_glimmer_vision** -- [MuseGlimmerVisionConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerVisionConfig) (model MuseGlimmerVisionConfig)
- **muse_spark** -- [MuseSparkConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkConfig) (model MuseSparkConfig)
- **muse_spark_audio** -- [MuseSparkAudioConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkAudioConfig) (model MuseSparkAudioConfig)
- **muse_spark_text** -- [MuseSparkTextConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkTextConfig) (model MuseSparkTextConfig)
- **muse_spark_vision** -- [MuseSparkVisionConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkVisionConfig) (model MuseSparkVisionConfig)
- **musicflamingo** -- [MusicFlamingoConfig](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoConfig) (model MusicFlamingoConfig)
- **musicgen** -- [MusicgenConfig](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenConfig) (model MusicgenConfig)
- **musicgen_decoder** -- [MusicgenDecoderConfig](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenDecoderConfig) (model MusicgenDecoderConfig)
- **musicgen_melody** -- [MusicgenMelodyConfig](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyConfig) (model MusicgenMelodyConfig)
- **musicgen_melody_decoder** -- [MusicgenMelodyDecoderConfig](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyDecoderConfig) (model MusicgenMelodyDecoderConfig)
- **mvp** -- [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig) (model MvpConfig)
- **nanochat** -- [NanoChatConfig](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatConfig) (model NanoChatConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeConfig) (model Nemotron3_5TranscribeConfig)
- **nemotron** -- [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig) (model NemotronConfig)
- **nemotron3_5_asr** -- [Nemotron3_5AsrConfig](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrConfig) (model Nemotron3_5AsrConfig)
- **nemotron3_5_transcribe_asr_audio** -- [Nemotron3_5TranscribeAsrAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAsrAudioConfig) (model Nemotron3_5TranscribeAsrAudioConfig)
- **nemotron3_5_transcribe_audio** -- [Nemotron3_5TranscribeAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAudioConfig) (model Nemotron3_5TranscribeAudioConfig)
- **nemotron3_5_transcribe_word_timestamp** -- [Nemotron3_5TranscribeWordTimestampConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeWordTimestampConfig) (model Nemotron3_5TranscribeWordTimestampConfig)
- **nemotron3_diarization** -- [Nemotron3DiarizationConfig](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationConfig) (model Nemotron3DiarizationConfig)
- **nemotron3_diarization_audio** -- [Nemotron3DiarizationAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationAudioConfig) (model Nemotron3DiarizationAudioConfig)
- **nemotron_asr_streaming** -- [NemotronAsrStreamingConfig](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingConfig) (model NemotronAsrStreamingConfig)
- **nemotron_asr_streaming_encoder** -- [NemotronAsrStreamingEncoderConfig](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingEncoderConfig) (model NemotronAsrStreamingEncoderConfig)
- **nemotron_h** -- [NemotronHConfig](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHConfig) (model NemotronHConfig)
- **nemotron_h_omni** -- [NemotronH_Omni_Reasoning_V3_Config](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3_Config) (model NemotronH_Omni_Reasoning_V3_Config)
- **neomme** -- [NeoMMEConfig](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEConfig) (model NeoMMEConfig)
- **neucodec** -- [NeuCodecConfig](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecConfig) (model NeuCodecConfig)
- **nllb-moe** -- [NllbMoeConfig](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeConfig) (model NllbMoeConfig)
- **nomic_bert** -- [NomicBertConfig](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertConfig) (model NomicBertConfig)
- **nougat** -- [NougatConfig](/docs/transformers/main/id/model_doc/nougat#transformers.NougatConfig) (model NougatConfig)
- **nystromformer** -- [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig) (model NystromformerConfig)
- **olmo** -- [OlmoConfig](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoConfig) (model OlmoConfig)
- **olmo2** -- [Olmo2Config](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Config) (model Olmo2Config)
- **olmo3** -- [Olmo3Config](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Config) (model Olmo3Config)
- **olmo_hybrid** -- [OlmoHybridConfig](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridConfig) (model OlmoHybridConfig)
- **olmoe** -- [OlmoeConfig](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeConfig) (model OlmoeConfig)
- **omdet-turbo** -- [OmDetTurboConfig](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboConfig) (model OmDetTurboConfig)
- **oneformer** -- [OneFormerConfig](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerConfig) (model OneFormerConfig)
- **openai-gpt** -- [OpenAIGPTConfig](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTConfig) (model OpenAIGPTConfig)
- **openai_privacy_filter** -- [OpenAIPrivacyFilterConfig](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterConfig) (model OpenAIPrivacyFilterConfig)
- **opt** -- [OPTConfig](/docs/transformers/main/id/model_doc/opt#transformers.OPTConfig) (model OPTConfig)
- **ovis2** -- [Ovis2Config](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Config) (model Ovis2Config)
- **ovis2_vision** -- [Ovis2VisionConfig](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2VisionConfig) (model Ovis2VisionConfig)
- **owlv2** -- [Owlv2Config](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Config) (model Owlv2Config)
- **owlv2_text_model** -- [Owlv2TextConfig](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2TextConfig) (model Owlv2TextConfig)
- **owlv2_vision_model** -- [Owlv2VisionConfig](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2VisionConfig) (model Owlv2VisionConfig)
- **owlvit** -- [OwlViTConfig](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTConfig) (model OwlViTConfig)
- **owlvit_text_model** -- [OwlViTTextConfig](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTTextConfig) (model OwlViTTextConfig)
- **owlvit_vision_model** -- [OwlViTVisionConfig](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTVisionConfig) (model OwlViTVisionConfig)
- **paddleocr_vl** -- [PaddleOCRVLConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLConfig) (model PaddleOCRVLConfig)
- **paddleocr_vl_text** -- [PaddleOCRTextConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRTextConfig) (model PaddleOCRTextConfig)
- **paddleocr_vl_vision** -- [PaddleOCRVisionConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVisionConfig) (model PaddleOCRVisionConfig)
- **paligemma** -- [PaliGemmaConfig](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaConfig) (model PaliGemmaConfig)
- **parakeet_ctc** -- [ParakeetCTCConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetCTCConfig) (model ParakeetCTCConfig)
- **parakeet_encoder** -- [ParakeetEncoderConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetEncoderConfig) (model ParakeetEncoderConfig)
- **parakeet_rnnt** -- [ParakeetRNNTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetRNNTConfig) (model ParakeetRNNTConfig)
- **parakeet_tdt** -- [ParakeetTDTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTDTConfig) (model ParakeetTDTConfig)
- **patchtsmixer** -- [PatchTSMixerConfig](/docs/transformers/main/id/model_doc/patchtsmixer#transformers.PatchTSMixerConfig) (model PatchTSMixerConfig)
- **patchtst** -- [PatchTSTConfig](/docs/transformers/main/id/model_doc/patchtst#transformers.PatchTSTConfig) (model PatchTSTConfig)
- **pe_audio** -- [PeAudioConfig](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioConfig) (model PeAudioConfig)
- **pe_audio_encoder** -- [PeAudioEncoderConfig](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioEncoderConfig) (model PeAudioEncoderConfig)
- **pe_audio_video** -- [PeAudioVideoConfig](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoConfig) (model PeAudioVideoConfig)
- **pe_audio_video_encoder** -- [PeAudioVideoEncoderConfig](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoEncoderConfig) (model PeAudioVideoEncoderConfig)
- **pe_video** -- [PeVideoConfig](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoConfig) (model PeVideoConfig)
- **pe_video_encoder** -- [PeVideoEncoderConfig](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoEncoderConfig) (model PeVideoEncoderConfig)
- **pegasus** -- [PegasusConfig](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusConfig) (model PegasusConfig)
- **pegasus_x** -- [PegasusXConfig](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXConfig) (model PegasusXConfig)
- **perceiver** -- [PerceiverConfig](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverConfig) (model PerceiverConfig)
- **perception_lm** -- [PerceptionLMConfig](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMConfig) (model PerceptionLMConfig)
- **persimmon** -- [PersimmonConfig](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonConfig) (model PersimmonConfig)
- **phi** -- [PhiConfig](/docs/transformers/main/id/model_doc/phi#transformers.PhiConfig) (model PhiConfig)
- **phi3** -- [Phi3Config](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Config) (model Phi3Config)
- **phi4_multimodal** -- [Phi4MultimodalConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalConfig) (model Phi4MultimodalConfig)
- **phi4_multimodal_audio** -- [Phi4MultimodalAudioConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalAudioConfig) (model Phi4MultimodalAudioConfig)
- **phi4_multimodal_vision** -- [Phi4MultimodalVisionConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalVisionConfig) (model Phi4MultimodalVisionConfig)
- **phimoe** -- [PhimoeConfig](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeConfig) (model PhimoeConfig)
- **pi0** -- [PI0Config](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Config) (model PI0Config)
- **pix2struct** -- [Pix2StructConfig](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructConfig) (model Pix2StructConfig)
- **pix2struct_text_model** -- [Pix2StructTextConfig](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructTextConfig) (model Pix2StructTextConfig)
- **pix2struct_vision_model** -- [Pix2StructVisionConfig](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructVisionConfig) (model Pix2StructVisionConfig)
- **pixio** -- [PixioConfig](/docs/transformers/main/id/model_doc/pixio#transformers.PixioConfig) (model PixioConfig)
- **pixtral** -- [PixtralVisionConfig](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralVisionConfig) (model PixtralVisionConfig)
- **plbart** -- [PLBartConfig](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartConfig) (model PLBartConfig)
- **pocket_tts** -- [PocketTTSConfig](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSConfig) (model PocketTTSConfig)
- **pocket_tts_audio_tokenizer** -- [PocketTTSAudioTokenizerConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerConfig) (model PocketTTSAudioTokenizerConfig)
- **pocket_tts_audio_tokenizer_decoder** -- [PocketTTSAudioTokenizerDecoderConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerDecoderConfig) (model PocketTTSAudioTokenizerDecoderConfig)
- **pocket_tts_audio_tokenizer_encoder** -- [PocketTTSAudioTokenizerEncoderConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerEncoderConfig) (model PocketTTSAudioTokenizerEncoderConfig)
- **poolformer** -- [PoolFormerConfig](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerConfig) (model PoolFormerConfig)
- **pop2piano** -- [Pop2PianoConfig](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoConfig) (model Pop2PianoConfig)
- **pp_chart2table** -- [PPChart2TableConfig](/docs/transformers/main/id/model_doc/pp_chart2table#transformers.PPChart2TableConfig) (model PPChart2TableConfig)
- **pp_chart2table_vision** -- [PPChart2TableVisionConfig](/docs/transformers/main/id/model_doc/pp_chart2table#transformers.PPChart2TableVisionConfig) (model PPChart2TableVisionConfig)
- **pp_doclayout_v2** -- [PPDocLayoutV2Config](/docs/transformers/main/id/model_doc/pp_doclayout_v2#transformers.PPDocLayoutV2Config) (model PPDocLayoutV2Config)
- **pp_doclayout_v3** -- [PPDocLayoutV3Config](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3Config) (model PPDocLayoutV3Config)
- **pp_doclayout_v4** -- [PPDocLayoutV4Config](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4Config) (model PPDocLayoutV4Config)
- **pp_formulanet** -- [PPFormulaNetConfig](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetConfig) (model PPFormulaNetConfig)
- **pp_formulanet_text** -- [PPFormulaNetTextConfig](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetTextConfig) (model PPFormulaNetTextConfig)
- **pp_formulanet_vision** -- [PPFormulaNetVisionConfig](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetVisionConfig) (model PPFormulaNetVisionConfig)
- **pp_lcnet** -- [PPLCNetConfig](/docs/transformers/main/id/model_doc/pp_lcnet#transformers.PPLCNetConfig) (model PPLCNetConfig)
- **pp_lcnet_v3** -- [PPLCNetV3Config](/docs/transformers/main/id/model_doc/pp_lcnet_v3#transformers.PPLCNetV3Config) (model PPLCNetV3Config)
- **pp_lcnet_v4** -- [PPLCNetV4Config](/docs/transformers/main/id/model_doc/pp_lcnet_v4#transformers.PPLCNetV4Config) (model PPLCNetV4Config)
- **pp_ocrv5_mobile_det** -- [PPOCRV5MobileDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_det#transformers.PPOCRV5MobileDetConfig) (model PPOCRV5MobileDetConfig)
- **pp_ocrv5_mobile_rec** -- [PPOCRV5MobileRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecConfig) (model PPOCRV5MobileRecConfig)
- **pp_ocrv5_server_det** -- [PPOCRV5ServerDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_server_det#transformers.PPOCRV5ServerDetConfig) (model PPOCRV5ServerDetConfig)
- **pp_ocrv5_server_rec** -- [PPOCRV5ServerRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecConfig) (model PPOCRV5ServerRecConfig)
- **pp_ocrv6_medium_det** -- [PPOCRV6MediumDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetConfig) (model PPOCRV6MediumDetConfig)
- **pp_ocrv6_small_det** -- [PPOCRV6SmallDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetConfig) (model PPOCRV6SmallDetConfig)
- **pp_ocrv6_small_rec** -- [PPOCRV6SmallRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecConfig) (model PPOCRV6SmallRecConfig)
- **pp_ocrv6_tiny_rec** -- [PPOCRV6TinyRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecConfig) (model PPOCRV6TinyRecConfig)
- **prompt_depth_anything** -- [PromptDepthAnythingConfig](/docs/transformers/main/id/model_doc/prompt_depth_anything#transformers.PromptDepthAnythingConfig) (model PromptDepthAnythingConfig)
- **prophetnet** -- [ProphetNetConfig](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetConfig) (model ProphetNetConfig)
- **pvt** -- [PvtConfig](/docs/transformers/main/id/model_doc/pvt#transformers.PvtConfig) (model PvtConfig)
- **pvt_v2** -- [PvtV2Config](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2Config) (model PvtV2Config)
- **qianfan_ocr** -- [QianfanOCRConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRConfig) (model QianfanOCRConfig)
- **qianfan_ocr_vision** -- [QianfanOCRVisionConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRVisionConfig) (model QianfanOCRVisionConfig)
- **qwen2** -- [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config) (model Qwen2Config)
- **qwen2_5_omni** -- [Qwen2_5OmniConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniConfig) (model Qwen2_5OmniConfig)
- **qwen2_5_omni_audio_encoder** -- [Qwen2_5OmniAudioEncoderConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniAudioEncoderConfig) (model Qwen2_5OmniAudioEncoderConfig)
- **qwen2_5_omni_bigvgan** -- [Qwen2_5OmniBigVGANConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniBigVGANConfig) (model Qwen2_5OmniBigVGANConfig)
- **qwen2_5_omni_dit** -- [Qwen2_5OmniDiTConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniDiTConfig) (model Qwen2_5OmniDiTConfig)
- **qwen2_5_omni_talker** -- [Qwen2_5OmniTalkerConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniTalkerConfig) (model Qwen2_5OmniTalkerConfig)
- **qwen2_5_omni_text** -- [Qwen2_5OmniTextConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniTextConfig) (model Qwen2_5OmniTextConfig)
- **qwen2_5_omni_thinker** -- [Qwen2_5OmniThinkerConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerConfig) (model Qwen2_5OmniThinkerConfig)
- **qwen2_5_omni_token2wav** -- [Qwen2_5OmniToken2WavConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniToken2WavConfig) (model Qwen2_5OmniToken2WavConfig)
- **qwen2_5_omni_vision_encoder** -- [Qwen2_5OmniVisionEncoderConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniVisionEncoderConfig) (model Qwen2_5OmniVisionEncoderConfig)
- **qwen2_5_vl** -- [Qwen2_5_VLConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLConfig) (model Qwen2_5_VLConfig)
- **qwen2_5_vl_text** -- [Qwen2_5_VLTextConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLTextConfig) (model Qwen2_5_VLTextConfig)
- **qwen2_5_vl_vision** -- [Qwen2_5_VLVisionConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLVisionConfig) (model Qwen2_5_VLVisionConfig)
- **qwen2_audio** -- [Qwen2AudioConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioConfig) (model Qwen2AudioConfig)
- **qwen2_audio_encoder** -- [Qwen2AudioEncoderConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioEncoderConfig) (model Qwen2AudioEncoderConfig)
- **qwen2_moe** -- [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig) (model Qwen2MoeConfig)
- **qwen2_vl** -- [Qwen2VLConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLConfig) (model Qwen2VLConfig)
- **qwen2_vl_text** -- [Qwen2VLTextConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLTextConfig) (model Qwen2VLTextConfig)
- **qwen2_vl_vision** -- [Qwen2VLVisionConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLVisionConfig) (model Qwen2VLVisionConfig)
- **qwen3** -- [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5MoeConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeConfig) (model Qwen3_5MoeConfig)
- **qwen3_5_moe_text** -- [Qwen3_5MoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeTextConfig) (model Qwen3_5MoeTextConfig)
- **qwen3_5_moe_vision** -- [Qwen3_5MoeVisionConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeVisionConfig) (model Qwen3_5MoeVisionConfig)
- **qwen3_5_text** -- [Qwen3_5TextConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextConfig) (model Qwen3_5TextConfig)
- **qwen3_5_vision** -- [Qwen3_5VisionConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5VisionConfig) (model Qwen3_5VisionConfig)
- **qwen3_asr** -- [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig) (model Qwen3ASRConfig)
- **qwen3_asr_encoder** -- [Qwen3ASREncoderConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASREncoderConfig) (model Qwen3ASREncoderConfig)
- **qwen3_moe** -- [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig) (model Qwen3NextConfig)
- **qwen3_omni_moe** -- [Qwen3OmniMoeConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeConfig) (model Qwen3OmniMoeConfig)
- **qwen3_omni_moe_audio_encoder** -- [Qwen3OmniMoeAudioEncoderConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeAudioEncoderConfig) (model Qwen3OmniMoeAudioEncoderConfig)
- **qwen3_omni_moe_talker_code_predictor** -- [Qwen3OmniMoeTalkerCodePredictorConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeTalkerCodePredictorConfig) (model Qwen3OmniMoeTalkerCodePredictorConfig)
- **qwen3_omni_moe_talker_text** -- [Qwen3OmniMoeTalkerTextConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeTalkerTextConfig) (model Qwen3OmniMoeTalkerTextConfig)
- **qwen3_omni_moe_text** -- [Qwen3OmniMoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeTextConfig) (model Qwen3OmniMoeTextConfig)
- **qwen3_omni_moe_thinker** -- [Qwen3OmniMoeThinkerConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerConfig) (model Qwen3OmniMoeThinkerConfig)
- **qwen3_omni_moe_vision_encoder** -- [Qwen3OmniMoeVisionEncoderConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeVisionEncoderConfig) (model Qwen3OmniMoeVisionEncoderConfig)
- **qwen3_vl** -- [Qwen3VLConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLConfig) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen3VLMoeConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeConfig) (model Qwen3VLMoeConfig)
- **qwen3_vl_moe_text** -- [Qwen3VLMoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeTextConfig) (model Qwen3VLMoeTextConfig)
- **qwen3_vl_moe_vision** -- [Qwen3VLMoeVisionConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeVisionConfig) (model Qwen3VLMoeVisionConfig)
- **qwen3_vl_text** -- [Qwen3VLTextConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLTextConfig) (model Qwen3VLTextConfig)
- **qwen3_vl_vision** -- [Qwen3VLVisionConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLVisionConfig) (model Qwen3VLVisionConfig)
- **qwen4_exp** -- [Qwen4ExpConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpConfig) (model Qwen4ExpConfig)
- **qwen4_exp_text** -- [Qwen4ExpTextConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpTextConfig) (model Qwen4ExpTextConfig)
- **qwen4_exp_vision** -- [Qwen4ExpVisionConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpVisionConfig) (model Qwen4ExpVisionConfig)
- **radio** -- [RadioConfig](/docs/transformers/main/id/model_doc/radio#transformers.RadioConfig) (model RadioConfig)
- **rag** -- [RagConfig](/docs/transformers/main/id/model_doc/rag#transformers.RagConfig) (model RagConfig)
- **recurrent_gemma** -- [RecurrentGemmaConfig](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaConfig) (model RecurrentGemmaConfig)
- **reformer** -- [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig) (model ReformerConfig)
- **regnet** -- [RegNetConfig](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetConfig) (model RegNetConfig)
- **rembert** -- [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig) (model RemBertConfig)
- **resnet** -- [ResNetConfig](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetConfig) (model ResNetConfig)
- **rf_detr** -- [RfDetrConfig](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrConfig) (model RfDetrConfig)
- **rf_detr_dinov2** -- [RfDetrDinov2Config](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrDinov2Config) (model RfDetrDinov2Config)
- **roberta** -- [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) (model RoCBertConfig)
- **roformer** -- [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig) (model RoFormerConfig)
- **rt_detr** -- [RTDetrConfig](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrConfig) (model RTDetrConfig)
- **rt_detr_resnet** -- [RTDetrResNetConfig](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrResNetConfig) (model RTDetrResNetConfig)
- **rt_detr_v2** -- [RTDetrV2Config](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2Config) (model RTDetrV2Config)
- **rwkv** -- [RwkvConfig](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvConfig) (model RwkvConfig)
- **sam** -- [SamConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamConfig) (model SamConfig)
- **sam2** -- [Sam2Config](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Config) (model Sam2Config)
- **sam2_hiera_det_model** -- [Sam2HieraDetConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2HieraDetConfig) (model Sam2HieraDetConfig)
- **sam2_mask_decoder** -- [Sam2MaskDecoderConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2MaskDecoderConfig) (model Sam2MaskDecoderConfig)
- **sam2_prompt_encoder** -- [Sam2PromptEncoderConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2PromptEncoderConfig) (model Sam2PromptEncoderConfig)
- **sam2_video** -- [Sam2VideoConfig](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoConfig) (model Sam2VideoConfig)
- **sam2_video_mask_decoder** -- [Sam2VideoMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoMaskDecoderConfig) (model Sam2VideoMaskDecoderConfig)
- **sam2_video_prompt_encoder** -- [Sam2VideoPromptEncoderConfig](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoPromptEncoderConfig) (model Sam2VideoPromptEncoderConfig)
- **sam2_vision_model** -- [Sam2VisionConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2VisionConfig) (model Sam2VisionConfig)
- **sam3** -- [Sam3Config](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Config) (model Sam3Config)
- **sam3_detr_decoder** -- [Sam3DETRDecoderConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3DETRDecoderConfig) (model Sam3DETRDecoderConfig)
- **sam3_detr_encoder** -- [Sam3DETREncoderConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3DETREncoderConfig) (model Sam3DETREncoderConfig)
- **sam3_geometry_encoder** -- [Sam3GeometryEncoderConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3GeometryEncoderConfig) (model Sam3GeometryEncoderConfig)
- **sam3_lite_text** -- [Sam3LiteTextConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextConfig) (model Sam3LiteTextConfig)
- **sam3_lite_text_detr_decoder** -- [Sam3LiteTextDETRDecoderConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextDETRDecoderConfig) (model Sam3LiteTextDETRDecoderConfig)
- **sam3_lite_text_detr_encoder** -- [Sam3LiteTextDETREncoderConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextDETREncoderConfig) (model Sam3LiteTextDETREncoderConfig)
- **sam3_lite_text_geometry_encoder** -- [Sam3LiteTextGeometryEncoderConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextGeometryEncoderConfig) (model Sam3LiteTextGeometryEncoderConfig)
- **sam3_lite_text_mask_decoder** -- [Sam3LiteTextMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextMaskDecoderConfig) (model Sam3LiteTextMaskDecoderConfig)
- **sam3_lite_text_text_model** -- [Sam3LiteTextTextConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextTextConfig) (model Sam3LiteTextTextConfig)
- **sam3_mask_decoder** -- [Sam3MaskDecoderConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3MaskDecoderConfig) (model Sam3MaskDecoderConfig)
- **sam3_tracker** -- [Sam3TrackerConfig](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerConfig) (model Sam3TrackerConfig)
- **sam3_tracker_mask_decoder** -- [Sam3TrackerMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerMaskDecoderConfig) (model Sam3TrackerMaskDecoderConfig)
- **sam3_tracker_prompt_encoder** -- [Sam3TrackerPromptEncoderConfig](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerPromptEncoderConfig) (model Sam3TrackerPromptEncoderConfig)
- **sam3_tracker_video** -- [Sam3TrackerVideoConfig](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoConfig) (model Sam3TrackerVideoConfig)
- **sam3_tracker_video_mask_decoder** -- [Sam3TrackerVideoMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoMaskDecoderConfig) (model Sam3TrackerVideoMaskDecoderConfig)
- **sam3_tracker_video_prompt_encoder** -- [Sam3TrackerVideoPromptEncoderConfig](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoPromptEncoderConfig) (model Sam3TrackerVideoPromptEncoderConfig)
- **sam3_video** -- [Sam3VideoConfig](/docs/transformers/main/id/model_doc/sam3_video#transformers.Sam3VideoConfig) (model Sam3VideoConfig)
- **sam3_vision_model** -- [Sam3VisionConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3VisionConfig) (model Sam3VisionConfig)
- **sam3_vit_model** -- [Sam3ViTConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3ViTConfig) (model Sam3ViTConfig)
- **sam_hq** -- [SamHQConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQConfig) (model SamHQConfig)
- **sam_hq_mask_decoder** -- [SamHQMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQMaskDecoderConfig) (model SamHQMaskDecoderConfig)
- **sam_hq_prompt_encoder** -- [SamHQPromptEncoderConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQPromptEncoderConfig) (model SamHQPromptEncoderConfig)
- **sam_hq_vision_model** -- [SamHQVisionConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQVisionConfig) (model SamHQVisionConfig)
- **sam_mask_decoder** -- [SamMaskDecoderConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamMaskDecoderConfig) (model SamMaskDecoderConfig)
- **sam_prompt_encoder** -- [SamPromptEncoderConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamPromptEncoderConfig) (model SamPromptEncoderConfig)
- **sam_vision_model** -- [SamVisionConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamVisionConfig) (model SamVisionConfig)
- **sapiens2** -- [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config) (model Sapiens2Config)
- **sapiens2_head** -- [Sapiens2HeadConfig](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2HeadConfig) (model Sapiens2HeadConfig)
- **seamless_m4t** -- [SeamlessM4TConfig](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TConfig) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4Tv2Config](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Config) (model SeamlessM4Tv2Config)
- **seed_oss** -- [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig) (model SeedOssConfig)
- **segformer** -- [SegformerConfig](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerConfig) (model SegformerConfig)
- **seggpt** -- [SegGptConfig](/docs/transformers/main/id/model_doc/seggpt#transformers.SegGptConfig) (model SegGptConfig)
- **sew** -- [SEWConfig](/docs/transformers/main/id/model_doc/sew#transformers.SEWConfig) (model SEWConfig)
- **sew-d** -- [SEWDConfig](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDConfig) (model SEWDConfig)
- **shieldgemma2** -- [ShieldGemma2Config](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2Config) (model ShieldGemma2Config)
- **siglip** -- [SiglipConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipConfig) (model SiglipConfig)
- **siglip2** -- [Siglip2Config](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Config) (model Siglip2Config)
- **siglip2_text_model** -- [Siglip2TextConfig](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2TextConfig) (model Siglip2TextConfig)
- **siglip2_vision_model** -- [Siglip2VisionConfig](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2VisionConfig) (model Siglip2VisionConfig)
- **siglip_text_model** -- [SiglipTextConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipTextConfig) (model SiglipTextConfig)
- **siglip_vision_model** -- [SiglipVisionConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipVisionConfig) (model SiglipVisionConfig)
- **slanet** -- [SLANetConfig](/docs/transformers/main/id/model_doc/slanet#transformers.SLANetConfig) (model SLANetConfig)
- **slanext** -- [SLANeXtConfig](/docs/transformers/main/id/model_doc/slanext#transformers.SLANeXtConfig) (model SLANeXtConfig)
- **slanext_vision** -- [SLANeXtVisionConfig](/docs/transformers/main/id/model_doc/slanext#transformers.SLANeXtVisionConfig) (model SLANeXtVisionConfig)
- **smollm3** -- [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config) (model SmolLM3Config)
- **smolvlm** -- [SmolVLMConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMConfig) (model SmolVLMConfig)
- **smolvlm_vision** -- [SmolVLMVisionConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMVisionConfig) (model SmolVLMVisionConfig)
- **solar_open** -- [SolarOpenConfig](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenConfig) (model SolarOpenConfig)
- **speech-encoder-decoder** -- [SpeechEncoderDecoderConfig](/docs/transformers/main/id/model_doc/speech-encoder-decoder#transformers.SpeechEncoderDecoderConfig) (model SpeechEncoderDecoderConfig)
- **speech_to_text** -- [Speech2TextConfig](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextConfig) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5Config](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Config) (model SpeechT5Config)
- **speecht5_hifigan** -- [SpeechT5HifiGanConfig](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5HifiGanConfig) (model SpeechT5HifiGanConfig)
- **splinter** -- [SplinterConfig](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterConfig) (model SplinterConfig)
- **squeezebert** -- [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig) (model SqueezeBertConfig)
- **stablelm** -- [StableLmConfig](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmConfig) (model StableLmConfig)
- **starcoder2** -- [Starcoder2Config](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Config) (model Starcoder2Config)
- **step3p5** -- [Step3p7TextConfig](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7TextConfig) (model Step3p7TextConfig)
- **step3p5_vision** -- [Step3p7VisionConfig](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7VisionConfig) (model Step3p7VisionConfig)
- **step3p7** -- [Step3p7Config](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Config) (model Step3p7Config)
- **superglue** -- [SuperGlueConfig](/docs/transformers/main/id/model_doc/superglue#transformers.SuperGlueConfig) (model SuperGlueConfig)
- **superpoint** -- [SuperPointConfig](/docs/transformers/main/id/model_doc/superpoint#transformers.SuperPointConfig) (model SuperPointConfig)
- **swiftformer** -- [SwiftFormerConfig](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerConfig) (model SwiftFormerConfig)
- **swin** -- [SwinConfig](/docs/transformers/main/id/model_doc/swin#transformers.SwinConfig) (model SwinConfig)
- **swin2sr** -- [Swin2SRConfig](/docs/transformers/main/id/model_doc/swin2sr#transformers.Swin2SRConfig) (model Swin2SRConfig)
- **swinv2** -- [Swinv2Config](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Config) (model Swinv2Config)
- **switch_transformers** -- [SwitchTransformersConfig](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersConfig) (model SwitchTransformersConfig)
- **t5** -- [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config) (model T5Config)
- **t5_gemma_module** -- [T5GemmaModuleConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaModuleConfig) (model T5GemmaModuleConfig)
- **t5gemma** -- [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) (model T5Gemma2Config)
- **t5gemma2_decoder** -- [T5Gemma2DecoderConfig](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2DecoderConfig) (model T5Gemma2DecoderConfig)
- **t5gemma2_encoder** -- [T5Gemma2EncoderConfig](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2EncoderConfig) (model T5Gemma2EncoderConfig)
- **t5gemma2_text** -- [T5Gemma2TextConfig](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2TextConfig) (model T5Gemma2TextConfig)
- **table-transformer** -- [TableTransformerConfig](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerConfig) (model TableTransformerConfig)
- **tapas** -- [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig) (model TapasConfig)
- **textnet** -- [TextNetConfig](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetConfig) (model TextNetConfig)
- **time_series_transformer** -- [TimeSeriesTransformerConfig](/docs/transformers/main/id/model_doc/time_series_transformer#transformers.TimeSeriesTransformerConfig) (model TimeSeriesTransformerConfig)
- **timesfm** -- [TimesFmConfig](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmConfig) (model TimesFmConfig)
- **timesfm2_5** -- [TimesFm2_5Config](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5Config) (model TimesFm2_5Config)
- **timesformer** -- [TimesformerConfig](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerConfig) (model TimesformerConfig)
- **timm_backbone** -- [TimmBackboneConfig](/docs/transformers/main/id/main_classes/backbones#transformers.TimmBackboneConfig) (model TimmBackboneConfig)
- **timm_wrapper** -- [TimmWrapperConfig](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperConfig) (model TimmWrapperConfig)
- **tipsv2** -- [Tipsv2Config](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Config) (model Tipsv2Config)
- **tipsv2_dpt** -- [Tipsv2DptConfig](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptConfig) (model Tipsv2DptConfig)
- **tipsv2_text_model** -- [Tipsv2TextConfig](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2TextConfig) (model Tipsv2TextConfig)
- **tipsv2_vision_model** -- [Tipsv2VisionConfig](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2VisionConfig) (model Tipsv2VisionConfig)
- **trocr** -- [TrOCRConfig](/docs/transformers/main/id/model_doc/trocr#transformers.TrOCRConfig) (model TrOCRConfig)
- **tvp** -- [TvpConfig](/docs/transformers/main/id/model_doc/tvp#transformers.TvpConfig) (model TvpConfig)
- **udop** -- [UdopConfig](/docs/transformers/main/id/model_doc/udop#transformers.UdopConfig) (model UdopConfig)
- **ultrabert** -- [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig) (model UltrabertConfig)
- **ultravbert** -- [UltraVBertConfig](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertConfig) (model UltraVBertConfig)
- **umt5** -- [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config) (model UMT5Config)
- **unispeech** -- [UniSpeechConfig](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechConfig) (model UniSpeechConfig)
- **unispeech-sat** -- [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig) (model UniSpeechSatConfig)
- **univnet** -- [UnivNetConfig](/docs/transformers/main/id/model_doc/univnet#transformers.UnivNetConfig) (model UnivNetConfig)
- **unlimited_ocr** -- [UnlimitedOcrConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrConfig) (model UnlimitedOcrConfig)
- **unlimited_ocr_sam_vision_model** -- [UnlimitedOcrSamVisionConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrSamVisionConfig) (model UnlimitedOcrSamVisionConfig)
- **unlimited_ocr_text** -- [UnlimitedOcrTextConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrTextConfig) (model UnlimitedOcrTextConfig)
- **unlimited_ocr_vision** -- [UnlimitedOcrVisionConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrVisionConfig) (model UnlimitedOcrVisionConfig)
- **unlimited_ocr_vision_encoder** -- [UnlimitedOcrVisionEncoderConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrVisionEncoderConfig) (model UnlimitedOcrVisionEncoderConfig)
- **upernet** -- [UperNetConfig](/docs/transformers/main/id/model_doc/upernet#transformers.UperNetConfig) (model UperNetConfig)
- **uvdoc** -- [UVDocConfig](/docs/transformers/main/id/model_doc/uvdoc#transformers.UVDocConfig) (model UVDocConfig)
- **uvdoc_backbone** -- [UVDocBackboneConfig](/docs/transformers/main/id/model_doc/uvdoc#transformers.UVDocBackboneConfig) (model UVDocBackboneConfig)
- **vaultgemma** -- [VaultGemmaConfig](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaConfig) (model VaultGemmaConfig)
- **vibevoice** -- [VibeVoiceConfig](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceConfig) (model VibeVoiceConfig)
- **vibevoice_acoustic_tokenizer** -- [VibeVoiceAcousticTokenizerConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerConfig) (model VibeVoiceAcousticTokenizerConfig)
- **vibevoice_acoustic_tokenizer_decoder** -- [VibeVoiceAcousticTokenizerDecoderConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerDecoderConfig) (model VibeVoiceAcousticTokenizerDecoderConfig)
- **vibevoice_acoustic_tokenizer_encoder** -- [VibeVoiceAcousticTokenizerEncoderConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerEncoderConfig) (model VibeVoiceAcousticTokenizerEncoderConfig)
- **vibevoice_asr** -- [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig) (model VibeVoiceAsrConfig)
- **vibevoice_diffusion** -- [VibeVoiceDiffusionHeadConfig](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceDiffusionHeadConfig) (model VibeVoiceDiffusionHeadConfig)
- **video_llama_3** -- [VideoLlama3Config](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Config) (model VideoLlama3Config)
- **video_llama_3_vision** -- [VideoLlama3VisionConfig](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3VisionConfig) (model VideoLlama3VisionConfig)
- **video_llava** -- [VideoLlavaConfig](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaConfig) (model VideoLlavaConfig)
- **videomae** -- [VideoMAEConfig](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEConfig) (model VideoMAEConfig)
- **videomt** -- [VideomtConfig](/docs/transformers/main/id/model_doc/videomt#transformers.VideomtConfig) (model VideomtConfig)
- **videoprism** -- [VideoPrismConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismConfig) (model VideoPrismConfig)
- **videoprism_text_model** -- [VideoPrismTextConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismTextConfig) (model VideoPrismTextConfig)
- **videoprism_vision_model** -- [VideoPrismVisionConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismVisionConfig) (model VideoPrismVisionConfig)
- **vilt** -- [ViltConfig](/docs/transformers/main/id/model_doc/vilt#transformers.ViltConfig) (model ViltConfig)
- **vipllava** -- [VipLlavaConfig](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaConfig) (model VipLlavaConfig)
- **vision-encoder-decoder** -- [VisionEncoderDecoderConfig](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderConfig) (model VisionEncoderDecoderConfig)
- **vision-text-dual-encoder** -- [VisionTextDualEncoderConfig](/docs/transformers/main/id/model_doc/vision-text-dual-encoder#transformers.VisionTextDualEncoderConfig) (model VisionTextDualEncoderConfig)
- **visual_bert** -- [VisualBertConfig](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertConfig) (model VisualBertConfig)
- **vit** -- [ViTConfig](/docs/transformers/main/id/model_doc/vit#transformers.ViTConfig) (model ViTConfig)
- **vit_mae** -- [ViTMAEConfig](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEConfig) (model ViTMAEConfig)
- **vit_msn** -- [ViTMSNConfig](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNConfig) (model ViTMSNConfig)
- **vitdet** -- [VitDetConfig](/docs/transformers/main/id/model_doc/vitdet#transformers.VitDetConfig) (model VitDetConfig)
- **vitmatte** -- [VitMatteConfig](/docs/transformers/main/id/model_doc/vitmatte#transformers.VitMatteConfig) (model VitMatteConfig)
- **vitpose** -- [VitPoseConfig](/docs/transformers/main/id/model_doc/vitpose#transformers.VitPoseConfig) (model VitPoseConfig)
- **vitpose_backbone** -- `VitPoseBackboneConfig` (model VitPoseBackboneConfig)
- **vits** -- [VitsConfig](/docs/transformers/main/id/model_doc/vits#transformers.VitsConfig) (model VitsConfig)
- **vivit** -- [VivitConfig](/docs/transformers/main/id/model_doc/vivit#transformers.VivitConfig) (model VivitConfig)
- **vjepa2** -- [VJEPA2Config](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2Config) (model VJEPA2Config)
- **voxtral** -- [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig) (model VoxtralConfig)
- **voxtral_encoder** -- [VoxtralEncoderConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralEncoderConfig) (model VoxtralEncoderConfig)
- **voxtral_realtime** -- [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig) (model VoxtralRealtimeConfig)
- **voxtral_realtime_encoder** -- [VoxtralRealtimeEncoderConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeEncoderConfig) (model VoxtralRealtimeEncoderConfig)
- **voxtral_realtime_text** -- [VoxtralRealtimeTextConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeTextConfig) (model VoxtralRealtimeTextConfig)
- **wav2vec2** -- [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig) (model WavLMConfig)
- **whisper** -- [WhisperConfig](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperConfig) (model WhisperConfig)
- **xclip** -- [XCLIPConfig](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPConfig) (model XCLIPConfig)
- **xclip_text_model** -- [XCLIPTextConfig](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPTextConfig) (model XCLIPTextConfig)
- **xclip_vision_model** -- [XCLIPVisionConfig](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPVisionConfig) (model XCLIPVisionConfig)
- **xcodec** -- [XcodecConfig](/docs/transformers/main/id/model_doc/xcodec#transformers.XcodecConfig) (model XcodecConfig)
- **xcodec2** -- [Xcodec2Config](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Config) (model Xcodec2Config)
- **xglm** -- [XGLMConfig](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMConfig) (model XGLMConfig)
- **xlm** -- [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) (model XLNetConfig)
- **xlstm** -- [xLSTMConfig](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMConfig) (model xLSTMConfig)
- **xmod** -- [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) (model XmodConfig)
- **yolos** -- [YolosConfig](/docs/transformers/main/id/model_doc/yolos#transformers.YolosConfig) (model YolosConfig)
- **yoso** -- [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig) (model YosoConfig)
- **youtu** -- [YoutuConfig](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuConfig) (model YoutuConfig)
- **zamba** -- [ZambaConfig](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaConfig) (model ZambaConfig)
- **zamba2** -- [Zamba2Config](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Config) (model Zamba2Config)
- **zaya** -- [ZayaConfig](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaConfig) (model ZayaConfig)
- **zoedepth** -- [ZoeDepthConfig](/docs/transformers/main/id/model_doc/zoedepth#transformers.ZoeDepthConfig) (model ZoeDepthConfig)

Contoh:

```python
>>> from transformers import AutoConfig

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-uncased")

>>> # Download configuration from huggingface.co (user-uploaded) and cache.
>>> config = AutoConfig.from_pretrained("dbmdz/bert-base-german-cased")

>>> # If configuration file is in a directory (e.g., was saved using *save_pretrained('./test/saved_model/')*).
>>> config = AutoConfig.from_pretrained("./test/bert_saved_model/")

>>> # Load a specific configuration file.
>>> config = AutoConfig.from_pretrained("./test/bert_saved_model/my_configuration.json")

>>> # Change some config attributes when loading a pretrained config.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-uncased", output_attentions=True, foo=False)
>>> config.output_attentions
True

>>> config, unused_kwargs = AutoConfig.from_pretrained(
...     "google-bert/bert-base-uncased", output_attentions=True, foo=False, return_unused_kwargs=True
... )
>>> config.output_attentions
True

>>> unused_kwargs
{'foo': False}
```

#### register[[transformers.AutoConfig.register]]

```python
register(model_type, config, exist_ok = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/configuration_auto.py#L455)

**Parameter:**

model_type (`str`) : Tipe model seperti "bert" atau "gpt".

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang akan didaftarkan.

Daftarkan konfigurasi baru untuk kelas ini.

## AutoTokenizer[[transformers.AutoTokenizer]]

#### transformers.AutoTokenizer[[transformers.AutoTokenizer]]

```python
transformers.AutoTokenizer()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/tokenization_auto.py#L656)

Ini adalah kelas tokenizer generik yang akan diinstansiasi sebagai salah satu kelas tokenizer dari pustaka ini saat dibuat dengan metode kelas [AutoTokenizer.from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoTokenizer.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoTokenizer.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/tokenization_auto.py#L670)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa: - String, yaitu *model id* dari tokenizer yang sudah ditentukan dan dihosting di dalam repo model di huggingface.co. - Path ke *direktori* yang berisi file kosakata yang diperlukan oleh tokenizer, misalnya yang disimpan menggunakan metode [save_pretrained()](/docs/transformers/main/id/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.save_pretrained), contohnya `./my_model_directory/`. - Path ke satu file kosakata yang disimpan jika dan hanya jika tokenizer hanya memerlukan satu file kosakata (seperti Bert atau XLNet), contohnya: `./my_model_directory/vocab.txt`. (Tidak berlaku untuk semua kelas turunan)

inputs (argumen posisional tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` Tokenizer.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Objek konfigurasi yang digunakan untuk menentukan kelas tokenizer yang akan diinstansiasi.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, defaultnya `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi serta menimpa versi yang di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

subfolder (`str`, *opsional*) : Jika file terkait berada di dalam subfolder dari repo model di huggingface.co (misalnya untuk facebook/rag-token-base), tentukan di sini.

tokenizer_type (`str`, *opsional*) : Jenis tokenizer yang akan dimuat.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

kwargs (argumen kata kunci tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` Tokenizer. Dapat digunakan untuk mengatur token khusus seperti `bos_token`, `eos_token`, `unk_token`, `sep_token`, `pad_token`, `cls_token`, `mask_token`, `additional_special_tokens`. Lihat parameter di `__init__()` untuk detail lebih lanjut.

Instansiasi salah satu kelas tokenizer dari pustaka ini dari kosakata model terlatih sebelumnya.

Kelas tokenizer yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **EvollaModel** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model EvollaConfig)
- **aimv2** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model Aimv2Config)
- **albert** -- [AlbertTokenizer](/docs/transformers/main/id/model_doc/albert#transformers.AlbertTokenizer) (model AlbertConfig)
- **align** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model AlignConfig)
- **aria** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model AriaConfig)
- **audioflamingo3** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model AudioFlamingo3Config)
- **aya_vision** -- [CohereTokenizer](/docs/transformers/main/id/model_doc/cohere#transformers.CohereTokenizer) (model AyaVisionConfig)
- **bark** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model BarkConfig)
- **bart** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model BartConfig)
- **beam** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model BeamConfig)
- **bert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model BertConfig)
- **bert-generation** -- [BertGenerationTokenizer](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationTokenizer) (model BertGenerationConfig)
- **big_bird** -- [BigBirdTokenizer](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdTokenizer) (model BigBirdConfig)
- **bigbird_pegasus** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model BigBirdPegasusConfig)
- **biogpt** -- [BioGptTokenizer](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptTokenizer) (model BioGptConfig)
- **blenderbot** -- [BlenderbotTokenizer](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotTokenizer) (model BlenderbotConfig)
- **blenderbot-small** -- [BlenderbotSmallTokenizer](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallTokenizer) (model BlenderbotSmallConfig)
- **blip** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model BlipConfig)
- **blip-2** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model Blip2Config)
- **bridgetower** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model BridgeTowerConfig)
- **bros** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model BrosConfig)
- **camembert** -- [CamembertTokenizer](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertTokenizer) (model CamembertConfig)
- **canary** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model CanaryConfig)
- **canine** -- [CanineTokenizer](/docs/transformers/main/id/model_doc/canine#transformers.CanineTokenizer) (model CanineConfig)
- **chinese_clip** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model ChineseCLIPConfig)
- **clap** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model ClapConfig)
- **clip** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model CLIPConfig)
- **clipseg** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model CLIPSegConfig)
- **clvp** -- [ClvpTokenizer](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpTokenizer) (model ClvpConfig)
- **codegen** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model CodeGenConfig)
- **cohere** -- [CohereTokenizer](/docs/transformers/main/id/model_doc/cohere#transformers.CohereTokenizer) (model CohereConfig)
- **cohere2** -- [CohereTokenizer](/docs/transformers/main/id/model_doc/cohere#transformers.CohereTokenizer) (model Cohere2Config)
- **cohere_asr** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model CohereAsrConfig)
- **colqwen2** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model ColQwen2Config)
- **convbert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model ConvBertConfig)
- **cosmos3_omni** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Cosmos3OmniConfig)
- **cpmant** -- [CpmAntTokenizer](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntTokenizer) (model CpmAntConfig)
- **ctrl** -- [CTRLTokenizer](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLTokenizer) (model CTRLConfig)
- **data2vec-audio** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model Data2VecAudioConfig)
- **data2vec-text** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model Data2VecTextConfig)
- **dbrx** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model DbrxConfig)
- **deberta** -- [DebertaTokenizer](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaTokenizer) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2Tokenizer](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Tokenizer) (model DebertaV2Config)
- **dia** -- [DiaTokenizer](/docs/transformers/main/id/model_doc/dia#transformers.DiaTokenizer) (model DiaConfig)
- **diffusion_gemma** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model DiffusionGemmaConfig)
- **distilbert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model DistilBertConfig)
- **dpr** -- [DPRQuestionEncoderTokenizer](/docs/transformers/main/id/model_doc/dpr#transformers.DPRQuestionEncoderTokenizer) (model DPRConfig)
- **electra** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model ElectraConfig)
- **emu3** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model Emu3Config)
- **ernie** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model ErnieConfig)
- **ernie4_5** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Ernie4_5Config)
- **ernie4_5_moe** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Ernie4_5_MoeConfig)
- **esm** -- [EsmTokenizer](/docs/transformers/main/id/model_doc/esm#transformers.EsmTokenizer) (model EsmConfig)
- **esmc** -- [EsmcTokenizer](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcTokenizer) (model EsmcConfig)
- **esmfold2** -- [EsmcTokenizer](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcTokenizer) (model EsmFold2Config)
- **falcon_mamba** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model FalconMambaConfig)
- **fastspeech2_conformer** -- `None` (model FastSpeech2ConformerConfig)
- **flaubert** -- [FlaubertTokenizer](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertTokenizer) (model FlaubertConfig)
- **flava** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model FlavaConfig)
- **flex_olmo** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model FlexOlmoConfig)
- **florence2** -- [BartTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model Florence2Config)
- **fnet** -- [FNetTokenizer](/docs/transformers/main/id/model_doc/fnet#transformers.FNetTokenizer) (model FNetConfig)
- **fsmt** -- [FSMTTokenizer](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTTokenizer) (model FSMTConfig)
- **fun_asr_nano** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model FunAsrNanoConfig)
- **funnel** -- [FunnelTokenizer](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelTokenizer) (model FunnelConfig)
- **fuyu** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model FuyuConfig)
- **gemma** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model GemmaConfig)
- **gemma2** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model Gemma2Config)
- **gemma3** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model Gemma3Config)
- **gemma3_text** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model Gemma3TextConfig)
- **gemma3n** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model Gemma3nConfig)
- **gemma3n_text** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model Gemma3nTextConfig)
- **git** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model GitConfig)
- **glm** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GlmConfig)
- **glm4** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Glm4Config)
- **glm4_moe** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Glm4MoeConfig)
- **glm4_moe_lite** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Glm4MoeLiteConfig)
- **glm4v** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Glm4vConfig)
- **glm4v_moe** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Glm4vMoeConfig)
- **glm_image** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GlmImageConfig)
- **glmasr** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GlmAsrConfig)
- **got_ocr2** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GotOcr2Config)
- **gpt-sw3** -- [GPTSw3Tokenizer](/docs/transformers/main/id/model_doc/gpt-sw3#transformers.GPTSw3Tokenizer) (model GPT2Config)
- **gpt2** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model GPT2Config)
- **gpt_bigcode** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model GPTNeoXConfig)
- **gpt_neox_japanese** -- [GPTNeoXJapaneseTokenizer](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseTokenizer) (model GPTNeoXJapaneseConfig)
- **gptj** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model GPTJConfig)
- **gradium_phonon_tts** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GradiumPhononTTSConfig)
- **granite** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GraniteConfig)
- **granite_speech5_ctc** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model GraniteSpeech5CTCConfig)
- **granitemoe** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GraniteMoeConfig)
- **granitemoehybrid** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GraniteMoeHybridConfig)
- **granitemoeshared** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model GraniteMoeSharedConfig)
- **grounding-dino** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model GroundingDinoConfig)
- **groupvit** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model GroupViTConfig)
- **gte** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model GteConfig)
- **hubert** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model HubertConfig)
- **hunyuan_vl** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model HunYuanVLConfig)
- **hyperclovax** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model HyperCLOVAXConfig)
- **hyperclovax_vision_v2** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model HyperCLOVAXVisionV2Config)
- **ibert** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model IBertConfig)
- **idefics** -- [LlamaTokenizer](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaTokenizer) (model IdeficsConfig)
- **idefics2** -- [LlamaTokenizer](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaTokenizer) (model Idefics2Config)
- **instructblip** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model InstructBlipConfig)
- **instructblipvideo** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model InstructBlipVideoConfig)
- **internvl** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model InternVLConfig)
- **jais2** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model Jais2Config)
- **jamba** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model JambaConfig)
- **jina_embeddings_v3** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model JinaEmbeddingsV3Config)
- **kimi_k25** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Kimi_K25Config)
- **kimi_linear** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model KimiLinearConfig)
- **kosmos-2** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Kosmos2Config)
- **lasr_ctc** -- [LasrTokenizer](/docs/transformers/main/id/model_doc/lasr#transformers.LasrTokenizer) (model LasrCTCConfig)
- **lasr_encoder** -- [LasrTokenizer](/docs/transformers/main/id/model_doc/lasr#transformers.LasrTokenizer) (model LasrEncoderConfig)
- **layoutlm** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2Tokenizer](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Tokenizer) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3Tokenizer](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Tokenizer) (model LayoutLMv3Config)
- **layoutxlm** -- [LayoutXLMTokenizer](/docs/transformers/main/id/model_doc/layoutxlm#transformers.LayoutXLMTokenizer) (model LayoutXLMConfig)
- **led** -- [LEDTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model LEDConfig)
- **lighton_ocr** -- [Qwen2TokenizerFast](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model LightOnOcrConfig)
- **lilt** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model LiltConfig)
- **llava** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model LlavaConfig)
- **llava_next** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model LlavaNextConfig)
- **longformer** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model LongformerConfig)
- **luke** -- [LukeTokenizer](/docs/transformers/main/id/model_doc/luke#transformers.LukeTokenizer) (model LukeConfig)
- **lxmert** -- [LxmertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model LxmertConfig)
- **m2m_100** -- [M2M100Tokenizer](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Tokenizer) (model M2M100Config)
- **mamba** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model MambaConfig)
- **mamba2** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model Mamba2Config)
- **marian** -- [MarianTokenizer](/docs/transformers/main/id/model_doc/marian#transformers.MarianTokenizer) (model MarianConfig)
- **markuplm** -- [MarkupLMTokenizer](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMTokenizer) (model MarkupLMConfig)
- **mbart** -- [MBartTokenizer](/docs/transformers/main/id/model_doc/mbart#transformers.MBartTokenizer) (model MBartConfig)
- **megatron-bert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model MegatronBertConfig)
- **metaclip_2** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model MetaClip2Config)
- **mgp-str** -- [MgpstrTokenizer](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrTokenizer) (model MgpstrConfig)
- **mimo_v2_flash** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiMoV2FlashConfig)
- **minicpm3** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiniCPM3Config)
- **minicpm_ocr** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiniCPMOcrConfig)
- **minicpmv4_6** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiniCPMV4_6Config)
- **minicpmv4_7** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiniCPMV4_7Config)
- **minimax_m2** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model MiniMaxM2Config)
- **ministral** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model MinistralConfig)
- **ministral3** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model Ministral3Config)
- **mistral** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model MistralConfig)
- **mistral3** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model Mistral3Config)
- **mistral_large4** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model MistralLarge4Config)
- **mixtral** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model MixtralConfig)
- **mm-grounding-dino** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model MMGroundingDinoConfig)
- **mobilebert** -- [MobileBertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model MobileBertConfig)
- **modernbert** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model ModernBertConfig)
- **modernbert-decoder** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model ModernBertDecoderConfig)
- **molmo2** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Molmo2Config)
- **moss_transcribe_diarize** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model MossTranscribeDiarizeConfig)
- **mpnet** -- [MPNetTokenizer](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetTokenizer) (model MPNetConfig)
- **mpt** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model MptConfig)
- **mra** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model MraConfig)
- **mt5** -- [T5Tokenizer](/docs/transformers/main/id/model_doc/t5#transformers.T5Tokenizer) (model MT5Config)
- **musicgen** -- [T5Tokenizer](/docs/transformers/main/id/model_doc/t5#transformers.T5Tokenizer) (model MusicgenConfig)
- **musicgen_melody** -- [T5Tokenizer](/docs/transformers/main/id/model_doc/t5#transformers.T5Tokenizer) (model MusicgenMelodyConfig)
- **mvp** -- [MvpTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model MvpConfig)
- **nemo_speechlm** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Nemotron3_5TranscribeConfig)
- **nemotron** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model NemotronConfig)
- **nemotron3_5_asr** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model Nemotron3_5AsrConfig)
- **nemotron_asr_streaming** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model NemotronAsrStreamingConfig)
- **nllb-moe** -- [NllbTokenizer](/docs/transformers/main/id/model_doc/nllb#transformers.NllbTokenizer) (model NllbMoeConfig)
- **nomic_bert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model NomicBertConfig)
- **nougat** -- [NougatTokenizer](/docs/transformers/main/id/model_doc/nougat#transformers.NougatTokenizer) (model NougatConfig)
- **nystromformer** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model NystromformerConfig)
- **olmo** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model OlmoConfig)
- **olmo2** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Olmo2Config)
- **olmo3** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Olmo3Config)
- **olmo_hybrid** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model OlmoHybridConfig)
- **olmoe** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model OlmoeConfig)
- **omdet-turbo** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model OmDetTurboConfig)
- **oneformer** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model OneFormerConfig)
- **openai-gpt** -- [OpenAIGPTTokenizer](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTTokenizer) (model OpenAIGPTConfig)
- **opt** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model OPTConfig)
- **ovis2** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Ovis2Config)
- **owlv2** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model Owlv2Config)
- **owlvit** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model OwlViTConfig)
- **paddleocr_vl** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model PaddleOCRVLConfig)
- **parakeet_ctc** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model ParakeetCTCConfig)
- **parakeet_rnnt** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model ParakeetRNNTConfig)
- **parakeet_tdt** -- [ParakeetTokenizer](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTokenizer) (model ParakeetTDTConfig)
- **pegasus** -- [PegasusTokenizer](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusTokenizer) (model PegasusConfig)
- **pegasus_x** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model PegasusXConfig)
- **perceiver** -- [PerceiverTokenizer](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverTokenizer) (model PerceiverConfig)
- **persimmon** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model PersimmonConfig)
- **phi** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model PhiConfig)
- **phi3** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Phi3Config)
- **phimoe** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model PhimoeConfig)
- **pix2struct** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model Pix2StructConfig)
- **pixtral** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model PixtralVisionConfig)
- **plbart** -- [PLBartTokenizer](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartTokenizer) (model PLBartConfig)
- **pocket_tts** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model PocketTTSConfig)
- **pp_formulanet** -- [NougatTokenizer](/docs/transformers/main/id/model_doc/nougat#transformers.NougatTokenizer) (model PPFormulaNetConfig)
- **prophetnet** -- [ProphetNetTokenizer](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetTokenizer) (model ProphetNetConfig)
- **qianfan_ocr** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model QianfanOCRConfig)
- **qwen2** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2Config)
- **qwen2_5_omni** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2_5OmniConfig)
- **qwen2_5_vl** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2_5_VLConfig)
- **qwen2_audio** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2AudioConfig)
- **qwen2_moe** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2MoeConfig)
- **qwen2_vl** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen2VLConfig)
- **qwen3** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5Tokenizer](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Tokenizer) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5Tokenizer](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Tokenizer) (model Qwen3_5MoeConfig)
- **qwen3_5_moe_text** -- [Qwen3_5Tokenizer](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Tokenizer) (model Qwen3_5MoeTextConfig)
- **qwen3_5_text** -- [Qwen3_5Tokenizer](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Tokenizer) (model Qwen3_5TextConfig)
- **qwen3_asr** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3ASRConfig)
- **qwen3_moe** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3NextConfig)
- **qwen3_omni_moe** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3OmniMoeConfig)
- **qwen3_vl** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen2Tokenizer](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model Qwen3VLMoeConfig)
- **qwen4_exp** -- [Qwen3_5Tokenizer](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Tokenizer) (model Qwen4ExpConfig)
- **rag** -- [RagTokenizer](/docs/transformers/main/id/model_doc/rag#transformers.RagTokenizer) (model RagConfig)
- **recurrent_gemma** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model RecurrentGemmaConfig)
- **reformer** -- [ReformerTokenizer](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerTokenizer) (model ReformerConfig)
- **rembert** -- [RemBertTokenizer](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertTokenizer) (model RemBertConfig)
- **roberta** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaTokenizer](/docs/transformers/main/id/model_doc/led#transformers.RobertaTokenizer) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertTokenizer](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertTokenizer) (model RoCBertConfig)
- **roformer** -- [RoFormerTokenizer](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerTokenizer) (model RoFormerConfig)
- **rwkv** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model RwkvConfig)
- **sam3** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model Sam3Config)
- **sam3_video** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model Sam3VideoConfig)
- **seamless_m4t** -- [SeamlessM4TTokenizer](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TTokenizer) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4TTokenizer](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TTokenizer) (model SeamlessM4Tv2Config)
- **shieldgemma2** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model ShieldGemma2Config)
- **siglip** -- [SiglipTokenizer](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipTokenizer) (model SiglipConfig)
- **siglip2** -- [Siglip2Tokenizer](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Tokenizer) (model Siglip2Config)
- **smolvlm** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model SmolVLMConfig)
- **speech_to_text** -- [Speech2TextTokenizer](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextTokenizer) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5Tokenizer](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Tokenizer) (model SpeechT5Config)
- **splinter** -- [SplinterTokenizer](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterTokenizer) (model SplinterConfig)
- **squeezebert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model SqueezeBertConfig)
- **stablelm** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model StableLmConfig)
- **starcoder2** -- [GPT2Tokenizer](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Tokenizer) (model Starcoder2Config)
- **switch_transformers** -- [T5Tokenizer](/docs/transformers/main/id/model_doc/t5#transformers.T5Tokenizer) (model SwitchTransformersConfig)
- **t5** -- [T5Tokenizer](/docs/transformers/main/id/model_doc/t5#transformers.T5Tokenizer) (model T5Config)
- **t5gemma** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model T5GemmaConfig)
- **tapas** -- [TapasTokenizer](/docs/transformers/main/id/model_doc/tapas#transformers.TapasTokenizer) (model TapasConfig)
- **tipsv2** -- [Tipsv2Tokenizer](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Tokenizer) (model Tipsv2Config)
- **trocr** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model TrOCRConfig)
- **tvp** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model TvpConfig)
- **udop** -- [UdopTokenizer](/docs/transformers/main/id/model_doc/udop#transformers.UdopTokenizer) (model UdopConfig)
- **umt5** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model UMT5Config)
- **unispeech** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model UniSpeechConfig)
- **unispeech-sat** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model UniSpeechSatConfig)
- **vibevoice** -- [Qwen2TokenizerFast](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Tokenizer) (model VibeVoiceConfig)
- **videoprism** -- [VideoPrismTokenizer](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismTokenizer) (model VideoPrismConfig)
- **vilt** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model ViltConfig)
- **vipllava** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model VipLlavaConfig)
- **visual_bert** -- [BertTokenizer](/docs/transformers/main/id/model_doc/lxmert#transformers.BertTokenizer) (model VisualBertConfig)
- **vits** -- [VitsTokenizer](/docs/transformers/main/id/model_doc/vits#transformers.VitsTokenizer) (model VitsConfig)
- **voxtral** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model VoxtralConfig)
- **voxtral_realtime** -- [MistralCommonBackend](/docs/transformers/main/id/model_doc/mistral#transformers.MistralCommonBackend) (model VoxtralRealtimeConfig)
- **wav2vec2** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2CTCTokenizer](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2CTCTokenizer) (model Wav2Vec2ConformerConfig)
- **whisper** -- [WhisperTokenizer](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperTokenizer) (model WhisperConfig)
- **xclip** -- [CLIPTokenizer](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTokenizer) (model XCLIPConfig)
- **xglm** -- [XGLMTokenizer](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMTokenizer) (model XGLMConfig)
- **xlm** -- [XLMTokenizer](/docs/transformers/main/id/model_doc/xlm#transformers.XLMTokenizer) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [TokenizersBackend](/docs/transformers/main/id/main_classes/tokenizer#transformers.TokenizersBackend) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetTokenizer](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetTokenizer) (model XLNetConfig)
- **xlstm** -- [GPTNeoXTokenizer](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXTokenizer) (model xLSTMConfig)
- **xmod** -- [XLMRobertaTokenizer](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaTokenizer) (model XmodConfig)
- **yoso** -- [AlbertTokenizer](/docs/transformers/main/id/model_doc/albert#transformers.AlbertTokenizer) (model YosoConfig)
- **zaya** -- [GemmaTokenizer](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaTokenizer) (model ZayaConfig)

Contoh:

```python
>>> from transformers import AutoTokenizer

>>> # Download vocabulary from huggingface.co and cache.
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")

>>> # Download vocabulary from huggingface.co (user-uploaded) and cache.
>>> tokenizer = AutoTokenizer.from_pretrained("dbmdz/bert-base-german-cased")

>>> # If vocabulary files are in a directory (e.g. tokenizer was saved using *save_pretrained('./test/saved_model/')*)
>>> # tokenizer = AutoTokenizer.from_pretrained("./test/bert_saved_model/")

>>> # Download vocabulary from huggingface.co and define model-specific arguments
>>> tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base", add_prefix_space=True)
```

#### register[[transformers.AutoTokenizer.register]]

```python
register(config_class, tokenizer_class = None, slow_tokenizer_class = None, fast_tokenizer_class = None, exist_ok = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/tokenization_auto.py#L1031)

**Parameter:**

config_class ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang sesuai dengan model yang akan didaftarkan.

tokenizer_class : Kelas tokenizer yang akan didaftarkan (V5 - parameter yang disarankan).

slow_tokenizer_class : (Tidak didukung lagi) Tokenizer lambat yang akan didaftarkan.

fast_tokenizer_class : (Tidak didukung lagi) Tokenizer cepat yang akan didaftarkan.

Daftarkan tokenizer baru dalam pemetaan ini.

## AutoFeatureExtractor[[transformers.AutoFeatureExtractor]]

#### transformers.AutoFeatureExtractor[[transformers.AutoFeatureExtractor]]

```python
transformers.AutoFeatureExtractor()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/feature_extraction_auto.py#L232)

Ini adalah kelas ekstraktor fitur generik yang akan diinstansiasi sebagai salah satu kelas ekstraktor fitur dari pustaka ini saat dibuat dengan metode kelas [AutoFeatureExtractor.from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoFeatureExtractor.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoFeatureExtractor.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/feature_extraction_auto.py#L246)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Ini bisa berupa:  - string, *model id* dari feature_extractor pretrained yang dihosting di dalam model repo di huggingface.co. - path ke *directory* yang berisi file feature extractor yang disimpan menggunakan metode [save_pretrained()](/docs/transformers/main/id/main_classes/feature_extractor#transformers.FeatureExtractionMixin.save_pretrained), misalnya `./my_model_directory/`. - path ke file JSON feature extractor yang disimpan, *file*, misalnya `./my_model_directory/preprocessor_config.json`.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat feature extractor model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, defaultnya `False`) : Apakah akan memaksa untuk (mengunduh ulang) file feature extractor dan menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.` Proxy digunakan pada setiap permintaan.

token (`str` atau *bool*, *opsional*) : Token yang akan digunakan sebagai otorisasi bearer HTTP untuk file jarak jauh. Jika `True`, akan menggunakan token yang dihasilkan saat menjalankan `hf auth login` (disimpan di `~/.huggingface`).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

return_unused_kwargs (`bool`, *opsional*, defaultnya `False`) : Jika `False`, fungsi ini hanya mengembalikan objek feature extractor akhir. Jika `True`, fungsi ini mengembalikan `Tuple(feature_extractor, unused_kwargs)` di mana *unused_kwargs* adalah kamus yang terdiri dari pasangan kunci/nilai yang kuncinya bukan merupakan atribut feature extractor: yaitu, bagian dari `kwargs` yang tidak digunakan untuk memperbarui `feature_extractor` dan diabaikan.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

kwargs (`dict[str, Any]`, *opsional*) : Nilai dalam kwargs untuk kunci apa pun yang merupakan atribut feature extractor akan digunakan untuk menimpa nilai yang dimuat. Perilaku terkait pasangan kunci/nilai yang kuncinya *bukan* atribut feature extractor dikendalikan oleh parameter kata kunci `return_unused_kwargs`.

Instansiasi salah satu kelas feature extractor dari pustaka ini dari kosakata model pretrained.

Kelas feature extractor yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pattern matching pada `pretrained_model_name_or_path`:

- **audio-spectrogram-transformer** -- [ASTFeatureExtractor](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTFeatureExtractor) (model ASTConfig)
- **audioflamingo3** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model AudioFlamingo3Config)
- **canary** -- [ParakeetFeatureExtractor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetFeatureExtractor) (model CanaryConfig)
- **clap** -- [ClapFeatureExtractor](/docs/transformers/main/id/model_doc/clap#transformers.ClapFeatureExtractor) (model ClapConfig)
- **clvp** -- [ClvpFeatureExtractor](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpFeatureExtractor) (model ClvpConfig)
- **cohere_asr** -- [CohereAsrFeatureExtractor](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrFeatureExtractor) (model CohereAsrConfig)
- **csm** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model CsmConfig)
- **dac** -- [DacFeatureExtractor](/docs/transformers/main/id/model_doc/dac#transformers.DacFeatureExtractor) (model DacConfig)
- **data2vec-audio** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model Data2VecAudioConfig)
- **dia** -- [DiaFeatureExtractor](/docs/transformers/main/id/model_doc/dia#transformers.DiaFeatureExtractor) (model DiaConfig)
- **embedding_gemma2** -- [Gemma4AudioFeatureExtractor](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioFeatureExtractor) (model EmbeddingGemma2Config)
- **encodec** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model EncodecConfig)
- **fun_asr_nano** -- [FunAsrNanoFeatureExtractor](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoFeatureExtractor) (model FunAsrNanoConfig)
- **gemma3n** -- [Gemma3nAudioFeatureExtractor](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nAudioFeatureExtractor) (model Gemma3nConfig)
- **gemma4** -- [Gemma4AudioFeatureExtractor](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioFeatureExtractor) (model Gemma4Config)
- **gemma4_unified** -- [Gemma4UnifiedAudioFeatureExtractor](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedAudioFeatureExtractor) (model Gemma4UnifiedConfig)
- **glmasr** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model GlmAsrConfig)
- **gradium_phonon_tts** -- [PocketTTSFeatureExtractor](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSFeatureExtractor) (model GradiumPhononTTSConfig)
- **granite_speech** -- [GraniteSpeechFeatureExtractor](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechFeatureExtractor) (model GraniteSpeechConfig)
- **granite_speech5_ctc** -- [GraniteSpeech5FeatureExtractor](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5FeatureExtractor) (model GraniteSpeech5CTCConfig)
- **granite_speech5_encoder** -- [GraniteSpeech5FeatureExtractor](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5FeatureExtractor) (model GraniteSpeech5EncoderConfig)
- **granite_speech_plus** -- [GraniteSpeechFeatureExtractor](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechFeatureExtractor) (model GraniteSpeechPlusConfig)
- **higgs_audio_v2_tokenizer** -- [DacFeatureExtractor](/docs/transformers/main/id/model_doc/dac#transformers.DacFeatureExtractor) (model HiggsAudioV2TokenizerConfig)
- **hubert** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model HubertConfig)
- **inkling_mm_model** -- [InklingFeatureExtractor](/docs/transformers/main/id/model_doc/inkling#transformers.InklingFeatureExtractor) (model InklingConfig)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextFeatureExtractor](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextFeatureExtractor) (model KyutaiSpeechToTextConfig)
- **lasr_ctc** -- [LasrFeatureExtractor](/docs/transformers/main/id/model_doc/lasr#transformers.LasrFeatureExtractor) (model LasrCTCConfig)
- **lasr_encoder** -- [LasrFeatureExtractor](/docs/transformers/main/id/model_doc/lasr#transformers.LasrFeatureExtractor) (model LasrEncoderConfig)
- **mimi** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model MimiConfig)
- **moonshine** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model MoonshineConfig)
- **moshi** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model MoshiConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeFeatureExtractor](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeFeatureExtractor) (model MossTranscribeDiarizeConfig)
- **muse_spark** -- [MuseSparkFeatureExtractor](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkFeatureExtractor) (model MuseSparkConfig)
- **musicgen** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model MusicgenConfig)
- **musicgen_melody** -- [MusicgenMelodyFeatureExtractor](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyFeatureExtractor) (model MusicgenMelodyConfig)
- **nemo_speechlm** -- [NemotronAsrStreamingFeatureExtractor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingFeatureExtractor) (model Nemotron3_5TranscribeConfig)
- **nemotron3_5_asr** -- [NemotronAsrStreamingFeatureExtractor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingFeatureExtractor) (model Nemotron3_5AsrConfig)
- **nemotron3_diarization** -- [NemotronAsrStreamingFeatureExtractor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingFeatureExtractor) (model Nemotron3DiarizationConfig)
- **nemotron_asr_streaming** -- [NemotronAsrStreamingFeatureExtractor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingFeatureExtractor) (model NemotronAsrStreamingConfig)
- **nemotron_asr_streaming_encoder** -- [NemotronAsrStreamingFeatureExtractor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingFeatureExtractor) (model NemotronAsrStreamingEncoderConfig)
- **neucodec** -- [NeuCodecFeatureExtractor](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecFeatureExtractor) (model NeuCodecConfig)
- **parakeet_ctc** -- [ParakeetFeatureExtractor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetFeatureExtractor) (model ParakeetCTCConfig)
- **parakeet_encoder** -- [ParakeetFeatureExtractor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetFeatureExtractor) (model ParakeetEncoderConfig)
- **parakeet_rnnt** -- [ParakeetFeatureExtractor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetFeatureExtractor) (model ParakeetRNNTConfig)
- **parakeet_tdt** -- [ParakeetFeatureExtractor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetFeatureExtractor) (model ParakeetTDTConfig)
- **pe_audio** -- [PeAudioFeatureExtractor](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioFeatureExtractor) (model PeAudioConfig)
- **pe_audio_video** -- [PeAudioFeatureExtractor](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioFeatureExtractor) (model PeAudioVideoConfig)
- **phi4_multimodal** -- [Phi4MultimodalFeatureExtractor](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalFeatureExtractor) (model Phi4MultimodalConfig)
- **pocket_tts** -- [PocketTTSFeatureExtractor](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSFeatureExtractor) (model PocketTTSConfig)
- **pocket_tts_audio_tokenizer** -- [EncodecFeatureExtractor](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecFeatureExtractor) (model PocketTTSAudioTokenizerConfig)
- **pop2piano** -- [Pop2PianoFeatureExtractor](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoFeatureExtractor) (model Pop2PianoConfig)
- **qwen2_5_omni** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model Qwen2_5OmniConfig)
- **qwen2_audio** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model Qwen2AudioConfig)
- **qwen3_asr** -- [Qwen3ASRFeatureExtractor](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRFeatureExtractor) (model Qwen3ASRConfig)
- **qwen3_omni_moe** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model Qwen3OmniMoeConfig)
- **seamless_m4t** -- [SeamlessM4TFeatureExtractor](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TFeatureExtractor) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4TFeatureExtractor](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TFeatureExtractor) (model SeamlessM4Tv2Config)
- **sew** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model SEWConfig)
- **sew-d** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model SEWDConfig)
- **speech_to_text** -- [Speech2TextFeatureExtractor](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextFeatureExtractor) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5FeatureExtractor](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5FeatureExtractor) (model SpeechT5Config)
- **unispeech** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model UniSpeechConfig)
- **unispeech-sat** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model UniSpeechSatConfig)
- **univnet** -- [UnivNetFeatureExtractor](/docs/transformers/main/id/model_doc/univnet#transformers.UnivNetFeatureExtractor) (model UnivNetConfig)
- **vibevoice** -- [VibeVoiceAcousticTokenizerFeatureExtractor](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerFeatureExtractor) (model VibeVoiceConfig)
- **vibevoice_acoustic_tokenizer** -- [VibeVoiceAcousticTokenizerFeatureExtractor](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerFeatureExtractor) (model VibeVoiceAcousticTokenizerConfig)
- **vibevoice_asr** -- [VibeVoiceAcousticTokenizerFeatureExtractor](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerFeatureExtractor) (model VibeVoiceAsrConfig)
- **voxtral** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeFeatureExtractor](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeFeatureExtractor) (model VoxtralRealtimeConfig)
- **wav2vec2** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [Wav2Vec2FeatureExtractor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2FeatureExtractor) (model WavLMConfig)
- **whisper** -- [WhisperFeatureExtractor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperFeatureExtractor) (model WhisperConfig)
- **xcodec** -- [DacFeatureExtractor](/docs/transformers/main/id/model_doc/dac#transformers.DacFeatureExtractor) (model XcodecConfig)
- **xcodec2** -- [Xcodec2FeatureExtractor](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2FeatureExtractor) (model Xcodec2Config)

Menyampaikan `token=True` diperlukan ketika Anda ingin menggunakan model privat.

Contoh:

```python
>>> from transformers import AutoFeatureExtractor

>>> # Download feature extractor from huggingface.co and cache.
>>> feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base-960h")

>>> # If feature extractor files are in a directory (e.g. feature extractor was saved using *save_pretrained('./test/saved_model/')*)
>>> # feature_extractor = AutoFeatureExtractor.from_pretrained("./test/saved_model/")
```

#### register[[transformers.AutoFeatureExtractor.register]]

```python
register(config_class, feature_extractor_class, exist_ok = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/feature_extraction_auto.py#L383)

**Parameter:**

config_class ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang sesuai dengan model yang akan didaftarkan.

feature_extractor_class (`FeatureExtractorMixin`) : Ekstraktor fitur yang akan didaftarkan.

Daftarkan fitur ekstraktor baru untuk kelas ini.

## AutoImageProcessor[[transformers.AutoImageProcessor]]

#### transformers.AutoImageProcessor[[transformers.AutoImageProcessor]]

```python
transformers.AutoImageProcessor()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/image_processing_auto.py#L497)

Ini adalah kelas pemroses gambar generik yang akan diinstansiasi sebagai salah satu kelas pemroses gambar dari pustaka ini saat dibuat dengan metode kelas [AutoImageProcessor.from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoImageProcessor.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoImageProcessor.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/image_processing_auto.py#L511)

**Parameter:**

pretrained_model_name_or_path (`str` or `os.PathLike`) : This can be either:  - a string, the *model id* of a pretrained image_processor hosted inside a model repo on huggingface.co. - a path to a *directory* containing a image processor file saved using the [save_pretrained()](/docs/transformers/main/id/internal/image_processing_utils#transformers.ImageProcessingMixin.save_pretrained) method, e.g., `./my_model_directory/`. - a path to a saved image processor JSON *file*, e.g., `./my_model_directory/preprocessor_config.json`.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat image processor model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa untuk (mengunduh ulang) file image processor dan menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.` Proxy digunakan pada setiap permintaan.

token (`str` atau *bool*, *opsional*) : Token yang akan digunakan sebagai otorisasi bearer HTTP untuk file jarak jauh. Jika `True`, akan menggunakan token yang dihasilkan saat menjalankan `hf auth login` (disimpan di `~/.huggingface`).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

use_fast (`bool`, *opsional*, defaultnya `False`) : **Usang**: Gunakan `backend="torchvision"` sebagai gantinya. Parameter ini dipertahankan untuk kompatibilitas mundur. Gunakan pemroses gambar berbasis torchvision yang cepat jika didukung untuk model tertentu. Jika pemroses gambar cepat tidak tersedia untuk model tertentu, pemroses gambar berbasis numpy biasa akan dikembalikan sebagai gantinya.

backend (`str`, *opsional*, defaultnya `None`) : Backend yang akan digunakan untuk pemrosesan gambar. Bisa berupa: - `None`: Secara otomatis memilih backend terbaik yang tersedia (torchvision jika tersedia, jika tidak pil) - `"torchvision"`: Gunakan backend Torchvision (dipercepat GPU, lebih cepat) - `"pil"`: Gunakan backend PIL (portabel, hanya CPU) - Nama backend kustom apa pun yang didaftarkan melalui metode `register()`

return_unused_kwargs (`bool`, *opsional*, default `False`) : Jika `False`, fungsi ini hanya mengembalikan objek image processor akhir. Jika `True`, fungsi ini mengembalikan `Tuple(image_processor, unused_kwargs)` di mana *unused_kwargs* adalah kamus yang terdiri dari pasangan kunci/nilai yang kuncinya bukan atribut image processor: yaitu, bagian dari `kwargs` yang tidak digunakan untuk memperbarui `image_processor` dan diabaikan.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

image_processor_filename (`str`, *opsional*, defaultnya `"config.json"`) : Nama file di direktori model yang digunakan untuk konfigurasi image processor.

kwargs (`dict[str, Any]`, *opsional*) : Nilai dalam kwargs untuk kunci apa pun yang merupakan atribut image processor akan digunakan untuk menimpa nilai yang dimuat. Perilaku terkait pasangan kunci/nilai yang kuncinya *bukan* atribut image processor dikendalikan oleh parameter kata kunci `return_unused_kwargs`.

Instansiasi salah satu kelas pemroses gambar dari pustaka ini dari kosakata model terlatih sebelumnya.

Kelas image processor yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pattern matching pada `pretrained_model_name_or_path`:

- **aimv2** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model Aimv2Config)
- **aimv2_vision_model** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model Aimv2VisionConfig)
- **align** -- `{'torchvision': 'EfficientNetImageProcessor', 'pil': 'EfficientNetImageProcessorPil'}` (model AlignConfig)
- **altclip** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model AltCLIPConfig)
- **aria** -- `{'pil': 'AriaImageProcessorPil', 'torchvision': 'AriaImageProcessor'}` (model AriaConfig)
- **aya_vision** -- `{'torchvision': 'GotOcr2ImageProcessor', 'pil': 'GotOcr2ImageProcessorPil'}` (model AyaVisionConfig)
- **beit** -- `{'pil': 'BeitImageProcessorPil', 'torchvision': 'BeitImageProcessor'}` (model BeitConfig)
- **bit** -- `{'pil': 'BitImageProcessorPil', 'torchvision': 'BitImageProcessor'}` (model BitConfig)
- **blip** -- `{'pil': 'BlipImageProcessorPil', 'torchvision': 'BlipImageProcessor'}` (model BlipConfig)
- **blip-2** -- `{'torchvision': 'BlipImageProcessor', 'pil': 'BlipImageProcessorPil'}` (model Blip2Config)
- **bridgetower** -- `{'pil': 'BridgeTowerImageProcessorPil', 'torchvision': 'BridgeTowerImageProcessor'}` (model BridgeTowerConfig)
- **chameleon** -- `{'pil': 'ChameleonImageProcessorPil', 'torchvision': 'ChameleonImageProcessor'}` (model ChameleonConfig)
- **chinese_clip** -- `{'pil': 'ChineseCLIPImageProcessorPil', 'torchvision': 'ChineseCLIPImageProcessor'}` (model ChineseCLIPConfig)
- **chmv2** -- `{'torchvision': 'CHMv2ImageProcessor'}` (model CHMv2Config)
- **clip** -- `{'pil': 'CLIPImageProcessorPil', 'torchvision': 'CLIPImageProcessor'}` (model CLIPConfig)
- **clipseg** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model CLIPSegConfig)
- **cohere2_vision** -- `{'torchvision': 'Cohere2VisionImageProcessor'}` (model Cohere2VisionConfig)
- **cohere_compass** -- `{'pil': 'CohereCompassImageProcessorPil', 'torchvision': 'CohereCompassImageProcessor'}` (model CohereCompassConfig)
- **colpali** -- `{'torchvision': 'SiglipImageProcessor', 'pil': 'SiglipImageProcessorPil'}` (model ColPaliConfig)
- **colqwen2** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model ColQwen2Config)
- **conditional_detr** -- `{'pil': 'ConditionalDetrImageProcessorPil', 'torchvision': 'ConditionalDetrImageProcessor'}` (model ConditionalDetrConfig)
- **convnext** -- `{'pil': 'ConvNextImageProcessorPil', 'torchvision': 'ConvNextImageProcessor'}` (model ConvNextConfig)
- **convnextv2** -- `{'torchvision': 'ConvNextImageProcessor', 'pil': 'ConvNextImageProcessorPil'}` (model ConvNextV2Config)
- **cosmos3_edge** -- `{'pil': 'Cosmos3EdgeImageProcessorPil', 'torchvision': 'Cosmos3EdgeImageProcessor'}` (model Cosmos3EdgeConfig)
- **cosmos3_omni** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Cosmos3OmniConfig)
- **cvt** -- `{'torchvision': 'ConvNextImageProcessor', 'pil': 'ConvNextImageProcessorPil'}` (model CvtConfig)
- **data2vec-vision** -- `{'torchvision': 'BeitImageProcessor', 'pil': 'BeitImageProcessorPil'}` (model Data2VecVisionConfig)
- **deepseek_ocr2** -- `{'pil': 'DeepseekOcr2ImageProcessorPil', 'torchvision': 'DeepseekOcr2ImageProcessor'}` (model DeepseekOcr2Config)
- **deepseek_vl** -- `{'pil': 'DeepseekVLImageProcessorPil', 'torchvision': 'DeepseekVLImageProcessor'}` (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- `{'pil': 'DeepseekVLHybridImageProcessorPil', 'torchvision': 'DeepseekVLHybridImageProcessor'}` (model DeepseekVLHybridConfig)
- **deformable_detr** -- `{'pil': 'DeformableDetrImageProcessorPil', 'torchvision': 'DeformableDetrImageProcessor'}` (model DeformableDetrConfig)
- **deimv2** -- `{'torchvision': 'RTDetrImageProcessor', 'pil': 'RTDetrImageProcessorPil'}` (model Deimv2Config)
- **deit** -- `{'pil': 'DeiTImageProcessorPil', 'torchvision': 'DeiTImageProcessor'}` (model DeiTConfig)
- **depth_anything** -- `{'torchvision': 'DPTImageProcessor', 'pil': 'DPTImageProcessorPil'}` (model DepthAnythingConfig)
- **depth_pro** -- `{'torchvision': 'DepthProImageProcessor'}` (model DepthProConfig)
- **detr** -- `{'pil': 'DetrImageProcessorPil', 'torchvision': 'DetrImageProcessor'}` (model DetrConfig)
- **dinat** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model DinatConfig)
- **dinov2** -- `{'torchvision': 'BitImageProcessor', 'pil': 'BitImageProcessorPil'}` (model Dinov2Config)
- **dinov3_vit** -- `{'torchvision': 'DINOv3ViTImageProcessor'}` (model DINOv3ViTConfig)
- **donut-swin** -- `{'torchvision': 'DonutImageProcessor', 'pil': 'DonutImageProcessorPil'}` (model DonutSwinConfig)
- **dpt** -- `{'pil': 'DPTImageProcessorPil', 'torchvision': 'DPTImageProcessor'}` (model DPTConfig)
- **edgetam** -- `{'torchvision': 'Sam2ImageProcessor'}` (model EdgeTamConfig)
- **efficientloftr** -- `{'pil': 'EfficientLoFTRImageProcessorPil', 'torchvision': 'EfficientLoFTRImageProcessor'}` (model EfficientLoFTRConfig)
- **efficientnet** -- `{'pil': 'EfficientNetImageProcessorPil', 'torchvision': 'EfficientNetImageProcessor'}` (model EfficientNetConfig)
- **embedding_gemma2** -- `{'pil': 'Gemma4ImageProcessorPil', 'torchvision': 'Gemma4ImageProcessor'}` (model EmbeddingGemma2Config)
- **emu3** -- `{'pil': 'Emu3ImageProcessor'}` (model Emu3Config)
- **eomt** -- `{'pil': 'EomtImageProcessorPil', 'torchvision': 'EomtImageProcessor'}` (model EomtConfig)
- **eomt_dinov3** -- `{'torchvision': 'EomtImageProcessor', 'pil': 'EomtImageProcessorPil'}` (model EomtDinov3Config)
- **ernie4_5_vl_moe** -- `{'pil': 'Ernie4_5_VLMoeImageProcessorPil', 'torchvision': 'Ernie4_5_VLMoeImageProcessor'}` (model Ernie4_5_VLMoeConfig)
- **exaone4_5** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Exaone4_5_Config)
- **flava** -- `{'pil': 'FlavaImageProcessorPil', 'torchvision': 'FlavaImageProcessor'}` (model FlavaConfig)
- **florence2** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model Florence2Config)
- **focalnet** -- `{'torchvision': 'BitImageProcessor', 'pil': 'BitImageProcessorPil'}` (model FocalNetConfig)
- **fuyu** -- `{'pil': 'FuyuImageProcessorPil', 'torchvision': 'FuyuImageProcessor'}` (model FuyuConfig)
- **gemma3** -- `{'pil': 'Gemma3ImageProcessorPil', 'torchvision': 'Gemma3ImageProcessor'}` (model Gemma3Config)
- **gemma3n** -- `{'torchvision': 'SiglipImageProcessor', 'pil': 'SiglipImageProcessorPil'}` (model Gemma3nConfig)
- **gemma4** -- `{'pil': 'Gemma4ImageProcessorPil', 'torchvision': 'Gemma4ImageProcessor'}` (model Gemma4Config)
- **gemma4_unified** -- `{'torchvision': 'Gemma4UnifiedImageProcessor'}` (model Gemma4UnifiedConfig)
- **git** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model GitConfig)
- **glm46v** -- `{'pil': 'Glm46VImageProcessorPil', 'torchvision': 'Glm46VImageProcessor'}` (model Glm46VConfig)
- **glm4v** -- `{'pil': 'Glm4vImageProcessorPil', 'torchvision': 'Glm4vImageProcessor'}` (model Glm4vConfig)
- **glm5_next** -- `{'pil': 'Glm5NextImageProcessorPil', 'torchvision': 'Glm5NextImageProcessor'}` (model Glm5NextConfig)
- **glm_image** -- `{'pil': 'GlmImageImageProcessorPil', 'torchvision': 'GlmImageImageProcessor'}` (model GlmImageConfig)
- **glmga** -- `{'pil': 'GlmgaImageProcessorPil', 'torchvision': 'GlmgaImageProcessor'}` (model GlmgaConfig)
- **glpn** -- `{'pil': 'GLPNImageProcessorPil', 'torchvision': 'GLPNImageProcessor'}` (model GLPNConfig)
- **got_ocr2** -- `{'pil': 'GotOcr2ImageProcessorPil', 'torchvision': 'GotOcr2ImageProcessor'}` (model GotOcr2Config)
- **granite4_vision** -- `{'torchvision': 'LlavaNextImageProcessor', 'pil': 'LlavaNextImageProcessorPil'}` (model Granite4VisionConfig)
- **grounding-dino** -- `{'pil': 'GroundingDinoImageProcessorPil', 'torchvision': 'GroundingDinoImageProcessor'}` (model GroundingDinoConfig)
- **groupvit** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model GroupViTConfig)
- **hiera** -- `{'torchvision': 'BitImageProcessor', 'pil': 'BitImageProcessorPil'}` (model HieraConfig)
- **hunyuan_vl** -- `{'pil': 'HunYuanVLImageProcessorPil', 'torchvision': 'HunYuanVLImageProcessor'}` (model HunYuanVLConfig)
- **hyperclovax_vision_v2** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model HyperCLOVAXVisionV2Config)
- **idefics** -- `{'pil': 'IdeficsImageProcessorPil', 'torchvision': 'IdeficsImageProcessor'}` (model IdeficsConfig)
- **idefics2** -- `{'pil': 'Idefics2ImageProcessorPil', 'torchvision': 'Idefics2ImageProcessor'}` (model Idefics2Config)
- **idefics3** -- `{'pil': 'Idefics3ImageProcessorPil', 'torchvision': 'Idefics3ImageProcessor'}` (model Idefics3Config)
- **ijepa** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model IJepaConfig)
- **imagegpt** -- `{'pil': 'ImageGPTImageProcessorPil', 'torchvision': 'ImageGPTImageProcessor'}` (model ImageGPTConfig)
- **inkling_mm_model** -- `{'torchvision': 'InklingImageProcessor'}` (model InklingConfig)
- **instructblip** -- `{'torchvision': 'BlipImageProcessor', 'pil': 'BlipImageProcessorPil'}` (model InstructBlipConfig)
- **internvl** -- `{'torchvision': 'GotOcr2ImageProcessor', 'pil': 'GotOcr2ImageProcessorPil'}` (model InternVLConfig)
- **janus** -- `{'pil': 'JanusImageProcessorPil', 'torchvision': 'JanusImageProcessor'}` (model JanusConfig)
- **kimi_k25** -- `{'torchvision': 'Kimi_K25ImageProcessor'}` (model Kimi_K25Config)
- **kosmos-2** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model Kosmos2Config)
- **kosmos-2.5** -- `{'torchvision': 'Kosmos2_5ImageProcessor', 'pil': 'Kosmos2_5ImageProcessorPil'}` (model Kosmos2_5Config)
- **layoutlmv2** -- `{'pil': 'LayoutLMv2ImageProcessorPil', 'torchvision': 'LayoutLMv2ImageProcessor'}` (model LayoutLMv2Config)
- **layoutlmv3** -- `{'pil': 'LayoutLMv3ImageProcessorPil', 'torchvision': 'LayoutLMv3ImageProcessor'}` (model LayoutLMv3Config)
- **layoutxlm** -- `{'torchvision': 'LayoutLMv2ImageProcessor', 'pil': 'LayoutLMv2ImageProcessorPil'}` (model LayoutXLMConfig)
- **levit** -- `{'pil': 'LevitImageProcessorPil', 'torchvision': 'LevitImageProcessor'}` (model LevitConfig)
- **lfm2_vl** -- `{'torchvision': 'Lfm2VlImageProcessor'}` (model Lfm2VlConfig)
- **lightglue** -- `{'pil': 'LightGlueImageProcessorPil', 'torchvision': 'LightGlueImageProcessor'}` (model LightGlueConfig)
- **lighton_ocr** -- `{'torchvision': 'PixtralImageProcessor', 'pil': 'PixtralImageProcessorPil'}` (model LightOnOcrConfig)
- **llama4** -- `{'torchvision': 'Llama4ImageProcessor'}` (model Llama4Config)
- **llava** -- `{'pil': 'LlavaImageProcessorPil', 'torchvision': 'LlavaImageProcessor'}` (model LlavaConfig)
- **llava_next** -- `{'pil': 'LlavaNextImageProcessorPil', 'torchvision': 'LlavaNextImageProcessor'}` (model LlavaNextConfig)
- **llava_next_video** -- `{'torchvision': 'LlavaNextImageProcessor', 'pil': 'LlavaNextImageProcessorPil'}` (model LlavaNextVideoConfig)
- **llava_onevision** -- `{'pil': 'LlavaOnevisionImageProcessorPil', 'torchvision': 'LlavaOnevisionImageProcessor'}` (model LlavaOnevisionConfig)
- **lw_detr** -- `{'torchvision': 'DeformableDetrImageProcessor', 'pil': 'DeformableDetrImageProcessorPil'}` (model LwDetrConfig)
- **mask2former** -- `{'pil': 'Mask2FormerImageProcessorPil', 'torchvision': 'Mask2FormerImageProcessor'}` (model Mask2FormerConfig)
- **maskformer** -- `{'pil': 'MaskFormerImageProcessorPil', 'torchvision': 'MaskFormerImageProcessor'}` (model MaskFormerConfig)
- **metaclip_2** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model MetaClip2Config)
- **mgp-str** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model MgpstrConfig)
- **minicpm_ocr** -- `{'torchvision': 'MiniCPMOcrImageProcessor'}` (model MiniCPMOcrConfig)
- **minicpmv4_6** -- `{'pil': 'MiniCPMV4_6ImageProcessorPil', 'torchvision': 'MiniCPMV4_6ImageProcessor'}` (model MiniCPMV4_6Config)
- **minicpmv4_7** -- `{'pil': 'MiniCPMV4_6ImageProcessorPil', 'torchvision': 'MiniCPMV4_6ImageProcessor'}` (model MiniCPMV4_7Config)
- **minimax_m3_vl** -- `{'torchvision': 'MiniMaxM3VLImageProcessor'}` (model MiniMaxM3VLConfig)
- **mistral3** -- `{'torchvision': 'PixtralImageProcessor', 'pil': 'PixtralImageProcessorPil'}` (model Mistral3Config)
- **mistral_large4** -- `{'torchvision': 'PixtralImageProcessor', 'pil': 'PixtralImageProcessorPil'}` (model MistralLarge4Config)
- **mlcd** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model MLCDVisionConfig)
- **mllama** -- `{'pil': 'MllamaImageProcessorPil', 'torchvision': 'MllamaImageProcessor'}` (model MllamaConfig)
- **mm-grounding-dino** -- `{'torchvision': 'GroundingDinoImageProcessor', 'pil': 'GroundingDinoImageProcessorPil'}` (model MMGroundingDinoConfig)
- **mobilenet_v1** -- `{'pil': 'MobileNetV1ImageProcessorPil', 'torchvision': 'MobileNetV1ImageProcessor'}` (model MobileNetV1Config)
- **mobilenet_v2** -- `{'pil': 'MobileNetV2ImageProcessorPil', 'torchvision': 'MobileNetV2ImageProcessor'}` (model MobileNetV2Config)
- **mobilevit** -- `{'pil': 'MobileViTImageProcessorPil', 'torchvision': 'MobileViTImageProcessor'}` (model MobileViTConfig)
- **mobilevitv2** -- `{'torchvision': 'MobileViTImageProcessor', 'pil': 'MobileViTImageProcessorPil'}` (model MobileViTV2Config)
- **molmo2** -- `{'torchvision': 'Molmo2ImageProcessor'}` (model Molmo2Config)
- **muse_glimmer** -- `{'torchvision': 'MuseGlimmerImageProcessor'}` (model MuseGlimmerConfig)
- **muse_spark** -- `{'torchvision': 'MuseGlimmerImageProcessor'}` (model MuseSparkConfig)
- **nemotron_h_omni** -- `{'torchvision': 'NemotronH_Omni_Reasoning_V3ImageProcessor'}` (model NemotronH_Omni_Reasoning_V3_Config)
- **neomme** -- `{'torchvision': 'NeoMMEImageProcessor'}` (model NeoMMEConfig)
- **nougat** -- `{'pil': 'NougatImageProcessorPil', 'torchvision': 'NougatImageProcessor'}` (model NougatConfig)
- **omdet-turbo** -- `{'torchvision': 'DetrImageProcessor', 'pil': 'DetrImageProcessorPil'}` (model OmDetTurboConfig)
- **oneformer** -- `{'pil': 'OneFormerImageProcessorPil', 'torchvision': 'OneFormerImageProcessor'}` (model OneFormerConfig)
- **ovis2** -- `{'pil': 'Ovis2ImageProcessorPil', 'torchvision': 'Ovis2ImageProcessor'}` (model Ovis2Config)
- **owlv2** -- `{'pil': 'Owlv2ImageProcessorPil', 'torchvision': 'Owlv2ImageProcessor'}` (model Owlv2Config)
- **owlvit** -- `{'pil': 'OwlViTImageProcessorPil', 'torchvision': 'OwlViTImageProcessor'}` (model OwlViTConfig)
- **paddleocr_vl** -- `{'pil': 'PaddleOCRVLImageProcessorPil', 'torchvision': 'PaddleOCRVLImageProcessor'}` (model PaddleOCRVLConfig)
- **paligemma** -- `{'torchvision': 'SiglipImageProcessor', 'pil': 'SiglipImageProcessorPil'}` (model PaliGemmaConfig)
- **perceiver** -- `{'pil': 'PerceiverImageProcessorPil', 'torchvision': 'PerceiverImageProcessor'}` (model PerceiverConfig)
- **perception_lm** -- `{'torchvision': 'PerceptionLMImageProcessor'}` (model PerceptionLMConfig)
- **phi4_multimodal** -- `{'torchvision': 'Phi4MultimodalImageProcessor'}` (model Phi4MultimodalConfig)
- **pi0** -- `{'torchvision': 'PI0ImageProcessor'}` (model PI0Config)
- **pix2struct** -- `{'pil': 'Pix2StructImageProcessorPil', 'torchvision': 'Pix2StructImageProcessor'}` (model Pix2StructConfig)
- **pixio** -- `{'torchvision': 'BitImageProcessor', 'pil': 'BitImageProcessorPil'}` (model PixioConfig)
- **pixtral** -- `{'pil': 'PixtralImageProcessorPil', 'torchvision': 'PixtralImageProcessor'}` (model PixtralVisionConfig)
- **poolformer** -- `{'pil': 'PoolFormerImageProcessorPil', 'torchvision': 'PoolFormerImageProcessor'}` (model PoolFormerConfig)
- **pp_chart2table** -- `{'pil': 'PPChart2TableImageProcessorPil', 'torchvision': 'PPChart2TableImageProcessor'}` (model PPChart2TableConfig)
- **pp_doclayout_v2** -- `{'torchvision': 'PPDocLayoutV2ImageProcessor'}` (model PPDocLayoutV2Config)
- **pp_doclayout_v3** -- `{'torchvision': 'PPDocLayoutV3ImageProcessor'}` (model PPDocLayoutV3Config)
- **pp_doclayout_v4** -- `{'torchvision': 'PPDocLayoutV4ImageProcessor'}` (model PPDocLayoutV4Config)
- **pp_formulanet** -- `{'torchvision': 'PPFormulaNetImageProcessor'}` (model PPFormulaNetConfig)
- **pp_lcnet** -- `{'torchvision': 'PPLCNetImageProcessor'}` (model PPLCNetConfig)
- **pp_ocrv5_mobile_det** -- `{'torchvision': 'PPOCRV5ServerDetImageProcessor'}` (model PPOCRV5MobileDetConfig)
- **pp_ocrv5_mobile_rec** -- `{'torchvision': 'PPOCRV5ServerRecImageProcessor'}` (model PPOCRV5MobileRecConfig)
- **pp_ocrv5_server_det** -- `{'torchvision': 'PPOCRV5ServerDetImageProcessor'}` (model PPOCRV5ServerDetConfig)
- **pp_ocrv5_server_rec** -- `{'torchvision': 'PPOCRV5ServerRecImageProcessor'}` (model PPOCRV5ServerRecConfig)
- **pp_ocrv6_medium_det** -- `{'torchvision': 'PPOCRV5ServerDetImageProcessor'}` (model PPOCRV6MediumDetConfig)
- **pp_ocrv6_small_det** -- `{'torchvision': 'PPOCRV5ServerDetImageProcessor'}` (model PPOCRV6SmallDetConfig)
- **pp_ocrv6_small_rec** -- `{'torchvision': 'PPOCRV6SmallRecImageProcessor'}` (model PPOCRV6SmallRecConfig)
- **pp_ocrv6_tiny_rec** -- `{'torchvision': 'PPOCRV6SmallRecImageProcessor'}` (model PPOCRV6TinyRecConfig)
- **prompt_depth_anything** -- `{'pil': 'PromptDepthAnythingImageProcessorPil', 'torchvision': 'PromptDepthAnythingImageProcessor'}` (model PromptDepthAnythingConfig)
- **pvt** -- `{'pil': 'PvtImageProcessorPil', 'torchvision': 'PvtImageProcessor'}` (model PvtConfig)
- **pvt_v2** -- `{'torchvision': 'PvtImageProcessor', 'pil': 'PvtImageProcessorPil'}` (model PvtV2Config)
- **qianfan_ocr** -- `{'torchvision': 'GotOcr2ImageProcessor', 'pil': 'GotOcr2ImageProcessorPil'}` (model QianfanOCRConfig)
- **qwen2_5_omni** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen2_5OmniConfig)
- **qwen2_5_vl** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen2_5_VLConfig)
- **qwen2_vl** -- `{'pil': 'Qwen2VLImageProcessorPil', 'torchvision': 'Qwen2VLImageProcessor'}` (model Qwen2VLConfig)
- **qwen3_5** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen3_5Config)
- **qwen3_5_moe** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen3_5MoeConfig)
- **qwen3_omni_moe** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen3OmniMoeConfig)
- **qwen3_vl** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen3VLConfig)
- **qwen4_exp** -- `{'torchvision': 'Qwen2VLImageProcessor', 'pil': 'Qwen2VLImageProcessorPil'}` (model Qwen4ExpConfig)
- **regnet** -- `{'torchvision': 'ConvNextImageProcessor', 'pil': 'ConvNextImageProcessorPil'}` (model RegNetConfig)
- **resnet** -- `{'torchvision': 'ConvNextImageProcessor', 'pil': 'ConvNextImageProcessorPil'}` (model ResNetConfig)
- **rf_detr** -- `{'torchvision': 'RfDetrImageProcessor'}` (model RfDetrConfig)
- **rt_detr** -- `{'pil': 'RTDetrImageProcessorPil', 'torchvision': 'RTDetrImageProcessor'}` (model RTDetrConfig)
- **sam** -- `{'pil': 'SamImageProcessorPil', 'torchvision': 'SamImageProcessor'}` (model SamConfig)
- **sam2** -- `{'torchvision': 'Sam2ImageProcessor'}` (model Sam2Config)
- **sam2_video** -- `{'torchvision': 'Sam2ImageProcessor'}` (model Sam2VideoConfig)
- **sam3** -- `{'torchvision': 'Sam3ImageProcessor'}` (model Sam3Config)
- **sam3_lite_text** -- `{'torchvision': 'Sam3ImageProcessor'}` (model Sam3LiteTextConfig)
- **sam3_tracker** -- `{'torchvision': 'Sam3ImageProcessor'}` (model Sam3TrackerConfig)
- **sam3_tracker_video** -- `{'torchvision': 'Sam3ImageProcessor'}` (model Sam3TrackerVideoConfig)
- **sam3_video** -- `{'torchvision': 'Sam3ImageProcessor'}` (model Sam3VideoConfig)
- **sam_hq** -- `{'torchvision': 'SamImageProcessor', 'pil': 'SamImageProcessorPil'}` (model SamHQConfig)
- **sapiens2** -- `{'torchvision': 'Sapiens2ImageProcessor'}` (model Sapiens2Config)
- **segformer** -- `{'pil': 'SegformerImageProcessorPil', 'torchvision': 'SegformerImageProcessor'}` (model SegformerConfig)
- **seggpt** -- `{'pil': 'SegGptImageProcessorPil', 'torchvision': 'SegGptImageProcessor'}` (model SegGptConfig)
- **shieldgemma2** -- `{'torchvision': 'Gemma3ImageProcessor', 'pil': 'Gemma3ImageProcessorPil'}` (model ShieldGemma2Config)
- **siglip** -- `{'pil': 'SiglipImageProcessorPil', 'torchvision': 'SiglipImageProcessor'}` (model SiglipConfig)
- **siglip2** -- `{'pil': 'Siglip2ImageProcessorPil', 'torchvision': 'Siglip2ImageProcessor'}` (model Siglip2Config)
- **slanet** -- `{'torchvision': 'SLANeXtImageProcessor'}` (model SLANetConfig)
- **slanext** -- `{'torchvision': 'SLANeXtImageProcessor'}` (model SLANeXtConfig)
- **smolvlm** -- `{'pil': 'SmolVLMImageProcessorPil', 'torchvision': 'SmolVLMImageProcessor'}` (model SmolVLMConfig)
- **step3p7** -- `{'torchvision': 'Step3p7ImageProcessor'}` (model Step3p7Config)
- **superglue** -- `{'pil': 'SuperGlueImageProcessorPil', 'torchvision': 'SuperGlueImageProcessor'}` (model SuperGlueConfig)
- **superpoint** -- `{'pil': 'SuperPointImageProcessorPil', 'torchvision': 'SuperPointImageProcessor'}` (model SuperPointConfig)
- **swiftformer** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model SwiftFormerConfig)
- **swin** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model SwinConfig)
- **swin2sr** -- `{'pil': 'Swin2SRImageProcessorPil', 'torchvision': 'Swin2SRImageProcessor'}` (model Swin2SRConfig)
- **swinv2** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model Swinv2Config)
- **t5gemma2** -- `{'torchvision': 'Gemma3ImageProcessor', 'pil': 'Gemma3ImageProcessorPil'}` (model T5Gemma2Config)
- **t5gemma2_encoder** -- `{'torchvision': 'Gemma3ImageProcessor', 'pil': 'Gemma3ImageProcessorPil'}` (model T5Gemma2EncoderConfig)
- **table-transformer** -- `{'torchvision': 'DetrImageProcessor', 'pil': 'DetrImageProcessorPil'}` (model TableTransformerConfig)
- **textnet** -- `{'pil': 'TextNetImageProcessorPil', 'torchvision': 'TextNetImageProcessor'}` (model TextNetConfig)
- **timesformer** -- `{'pil': 'VideoMAEImageProcessorPil', 'torchvision': 'VideoMAEImageProcessor'}` (model TimesformerConfig)
- **timm_wrapper** -- `{'pil': 'TimmWrapperImageProcessor'}` (model TimmWrapperConfig)
- **tipsv2** -- `{'torchvision': 'Tipsv2ImageProcessor'}` (model Tipsv2Config)
- **tipsv2_dpt** -- `{'torchvision': 'Tipsv2DptImageProcessor'}` (model Tipsv2DptConfig)
- **trocr** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model TrOCRConfig)
- **tvp** -- `{'pil': 'TvpImageProcessorPil', 'torchvision': 'TvpImageProcessor'}` (model TvpConfig)
- **udop** -- `{'torchvision': 'LayoutLMv3ImageProcessor', 'pil': 'LayoutLMv3ImageProcessorPil'}` (model UdopConfig)
- **ultravbert** -- `{'torchvision': 'UltraVBertImageProcessor'}` (model UltraVBertConfig)
- **unlimited_ocr** -- `{'torchvision': 'UnlimitedOcrImageProcessor'}` (model UnlimitedOcrConfig)
- **upernet** -- `{'torchvision': 'SegformerImageProcessor', 'pil': 'SegformerImageProcessorPil'}` (model UperNetConfig)
- **uvdoc** -- `{'torchvision': 'UVDocImageProcessor'}` (model UVDocConfig)
- **video_llama_3** -- `{'pil': 'VideoLlama3ImageProcessorPil', 'torchvision': 'VideoLlama3ImageProcessor'}` (model VideoLlama3Config)
- **video_llava** -- `{'pil': 'VideoLlavaImageProcessor'}` (model VideoLlavaConfig)
- **videomae** -- `{'pil': 'VideoMAEImageProcessorPil', 'torchvision': 'VideoMAEImageProcessor'}` (model VideoMAEConfig)
- **vilt** -- `{'pil': 'ViltImageProcessorPil', 'torchvision': 'ViltImageProcessor'}` (model ViltConfig)
- **vipllava** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model VipLlavaConfig)
- **vit** -- `{'pil': 'ViTImageProcessorPil', 'torchvision': 'ViTImageProcessor'}` (model ViTConfig)
- **vit_mae** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model ViTMAEConfig)
- **vit_msn** -- `{'torchvision': 'ViTImageProcessor', 'pil': 'ViTImageProcessorPil'}` (model ViTMSNConfig)
- **vitmatte** -- `{'pil': 'VitMatteImageProcessorPil', 'torchvision': 'VitMatteImageProcessor'}` (model VitMatteConfig)
- **vitpose** -- `{'pil': 'VitPoseImageProcessorPil', 'torchvision': 'VitPoseImageProcessor'}` (model VitPoseConfig)
- **vivit** -- `{'torchvision': 'VivitImageProcessor'}` (model VivitConfig)
- **xclip** -- `{'torchvision': 'CLIPImageProcessor', 'pil': 'CLIPImageProcessorPil'}` (model XCLIPConfig)
- **yolos** -- `{'pil': 'YolosImageProcessorPil', 'torchvision': 'YolosImageProcessor'}` (model YolosConfig)
- **zoedepth** -- `{'pil': 'ZoeDepthImageProcessorPil', 'torchvision': 'ZoeDepthImageProcessor'}` (model ZoeDepthConfig)

Menyampaikan `token=True` diperlukan ketika Anda ingin menggunakan model privat.

Contoh:

```python
>>> from transformers import AutoImageProcessor

>>> # Download image processor from huggingface.co and cache.
>>> image_processor = AutoImageProcessor.from_pretrained("google/vit-base-patch16-224-in21k")

>>> # If image processor files are in a directory (e.g. image processor was saved using *save_pretrained('./test/saved_model/')*)
>>> # image_processor = AutoImageProcessor.from_pretrained("./test/saved_model/")
```

#### register[[transformers.AutoImageProcessor.register]]

```python
register(config_class, slow_image_processor_class: type | None = None, fast_image_processor_class: type | None = None, image_processor_classes: dict[str, type] | None = None, exist_ok: bool = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/image_processing_auto.py#L725)

**Parameter:**

config_class ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang sesuai dengan model yang akan didaftarkan.

slow_image_processor_class (`type`, *opsional*) : Kelas pemroses gambar backend PIL (tidak didukung lagi, gunakan `image_processor_classes={"pil": ...}`).

fast_image_processor_class (`type`, *opsional*) : Kelas pemroses gambar backend Torchvision (tidak didukung lagi, gunakan `image_processor_classes={"torchvision": ...}`).

image_processor_classes (`dict[str, type]`, *opsional*) : Kamus yang memetakan nama backend ke kelas image processor. Memungkinkan pendaftaran backend kustom. Contoh: `{"pil": MyPilProcessor, "torchvision": MyTorchvisionProcessor, "custom": MyCustomProcessor}`

exist_ok (`bool`, *opsional*, defaultnya `False`) : Jika `True`, izinkan penulisan ulang pendaftaran yang sudah ada.

Daftarkan pemroses gambar baru untuk kelas ini.

## AutoVideoProcessor[[transformers.AutoVideoProcessor]]

Lihat [Backend pemroses video](../main_classes/video_processor#video-processor-backends) untuk pemilihan backend, pendaftaran kustom, serta penyimpanan dan pemuatan ulang.

#### transformers.AutoVideoProcessor[[transformers.AutoVideoProcessor]]

```python
transformers.AutoVideoProcessor()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/video_processing_auto.py#L264)

Ini adalah kelas pemroses video generik yang akan diinstansiasi sebagai salah satu kelas pemroses video dari pustaka ini saat dibuat dengan metode kelas [AutoVideoProcessor.from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoVideoProcessor.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoVideoProcessor.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/video_processing_auto.py#L278)

**Parameter:**

pretrained_model_name_or_path (`str` or `os.PathLike`) : This can be either:  - a string, the *model id* of a pretrained video_processor hosted inside a model repo on huggingface.co. - a path to a *directory* containing a video processor file saved using the [save_pretrained()](/docs/transformers/main/id/main_classes/video_processor#transformers.VideoProcessorMixin.save_pretrained) method, e.g., `./my_model_directory/`. - a path to a saved video processor JSON *file*, e.g., `./my_model_directory/preprocessor_config.json`.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat video processor model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, defaultnya `False`) : Apakah akan memaksa untuk (mengunduh ulang) file video processor dan menimpa versi yang di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.` Proxy digunakan pada setiap permintaan.

token (`str` atau *bool*, *opsional*) : Token yang akan digunakan sebagai otorisasi bearer HTTP untuk file jarak jauh. Jika `True`, akan menggunakan token yang dihasilkan saat menjalankan `hf auth login` (disimpan di `~/.huggingface`).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

return_unused_kwargs (`bool`, *opsional*, defaultnya `False`) : Jika `False`, maka fungsi ini hanya mengembalikan objek video processor akhir. Jika `True`, maka fungsi ini mengembalikan `Tuple(video_processor, unused_kwargs)` di mana *unused_kwargs* adalah kamus yang terdiri dari pasangan kunci/nilai yang kuncinya bukan atribut video processor: yaitu, bagian dari `kwargs` yang tidak digunakan untuk memperbarui `video_processor` dan diabaikan.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

kwargs (`dict[str, Any]`, *opsional*) : Nilai dalam kwargs dari kunci mana pun yang merupakan atribut video processor akan digunakan untuk menimpa nilai yang dimuat. Perilaku terkait pasangan kunci/nilai yang kuncinya *bukan* atribut video processor dikendalikan oleh parameter kata kunci `return_unused_kwargs`.

Instansiasi salah satu kelas pemroses video dari pustaka ini dari kosakata model terlatih sebelumnya.

Kelas video processor yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **cohere_compass** -- `{'torchvision': 'CohereCompassVideoProcessor'}` (model CohereCompassConfig)
- **cosmos3_edge** -- `{'torchvision': 'Cosmos3EdgeVideoProcessor'}` (model Cosmos3EdgeConfig)
- **cosmos3_omni** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Cosmos3OmniConfig)
- **embedding_gemma2** -- `{'torchvision': 'EmbeddingGemma2VideoProcessor'}` (model EmbeddingGemma2Config)
- **ernie4_5_vl_moe** -- `{'torchvision': 'Ernie4_5_VLMoeVideoProcessor'}` (model Ernie4_5_VLMoeConfig)
- **exaone4_5** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model Exaone4_5_Config)
- **gemma4** -- `{'torchvision': 'Gemma4VideoProcessor'}` (model Gemma4Config)
- **gemma4_unified** -- `{'torchvision': 'Gemma4UnifiedVideoProcessor'}` (model Gemma4UnifiedConfig)
- **glm46v** -- `{'torchvision': 'Glm46VVideoProcessor'}` (model Glm46VConfig)
- **glm4v** -- `{'torchvision': 'Glm4vVideoProcessor'}` (model Glm4vConfig)
- **glm5_next** -- `{'torchvision': 'Glm5NextVideoProcessor'}` (model Glm5NextConfig)
- **glmga** -- `{'torchvision': 'GlmgaVideoProcessor'}` (model GlmgaConfig)
- **hyperclovax_vision_v2** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model HyperCLOVAXVisionV2Config)
- **instructblip** -- `{'torchvision': 'InstructBlipVideoVideoProcessor'}` (model InstructBlipConfig)
- **instructblipvideo** -- `{'torchvision': 'InstructBlipVideoVideoProcessor'}` (model InstructBlipVideoConfig)
- **internvl** -- `{'torchvision': 'InternVLVideoProcessor'}` (model InternVLConfig)
- **kimi_k25** -- `{'torchvision': 'Kimi_K25VideoProcessor'}` (model Kimi_K25Config)
- **llava_next_video** -- `{'torchvision': 'LlavaNextVideoVideoProcessor'}` (model LlavaNextVideoConfig)
- **llava_onevision** -- `{'torchvision': 'LlavaOnevisionVideoProcessor'}` (model LlavaOnevisionConfig)
- **minicpm_ocr** -- `{'torchvision': 'MiniCPMOcrVideoProcessor'}` (model MiniCPMOcrConfig)
- **minicpmv4_6** -- `{'torchvision': 'MiniCPMV4_6VideoProcessor'}` (model MiniCPMV4_6Config)
- **minicpmv4_7** -- `{'torchvision': 'MiniCPMV4_6VideoProcessor'}` (model MiniCPMV4_7Config)
- **minimax_m3_vl** -- `{'torchvision': 'MiniMaxM3VLVideoProcessor'}` (model MiniMaxM3VLConfig)
- **molmo2** -- `{'torchvision': 'Molmo2VideoProcessor'}` (model Molmo2Config)
- **muse_glimmer** -- `{'torchvision': 'MuseGlimmerVideoProcessor'}` (model MuseGlimmerConfig)
- **muse_spark** -- `{'torchvision': 'MuseGlimmerVideoProcessor'}` (model MuseSparkConfig)
- **nemotron_h_omni** -- `{'torchvision': 'NemotronH_Omni_Reasoning_V3VideoProcessor'}` (model NemotronH_Omni_Reasoning_V3_Config)
- **pe_audio_video** -- `{'torchvision': 'PeVideoVideoProcessor'}` (model PeAudioVideoConfig)
- **pe_video** -- `{'torchvision': 'PeVideoVideoProcessor'}` (model PeVideoConfig)
- **perception_lm** -- `{'torchvision': 'PerceptionLMVideoProcessor'}` (model PerceptionLMConfig)
- **qwen2_5_omni** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model Qwen2_5OmniConfig)
- **qwen2_5_vl** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model Qwen2_5_VLConfig)
- **qwen2_vl** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model Qwen2VLConfig)
- **qwen3_5** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Qwen3_5Config)
- **qwen3_5_moe** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Qwen3_5MoeConfig)
- **qwen3_omni_moe** -- `{'torchvision': 'Qwen2VLVideoProcessor'}` (model Qwen3OmniMoeConfig)
- **qwen3_vl** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Qwen3VLConfig)
- **qwen3_vl_moe** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Qwen3VLMoeConfig)
- **qwen4_exp** -- `{'torchvision': 'Qwen3VLVideoProcessor'}` (model Qwen4ExpConfig)
- **sam2_video** -- `{'torchvision': 'Sam2VideoVideoProcessor'}` (model Sam2VideoConfig)
- **smolvlm** -- `{'torchvision': 'SmolVLMVideoProcessor'}` (model SmolVLMConfig)
- **timesformer** -- `{'torchvision': 'TimesformerVideoProcessor'}` (model TimesformerConfig)
- **video_llama_3** -- `{'torchvision': 'VideoLlama3VideoProcessor'}` (model VideoLlama3Config)
- **video_llava** -- `{'torchvision': 'VideoLlavaVideoProcessor'}` (model VideoLlavaConfig)
- **videomae** -- `{'torchvision': 'VideoMAEVideoProcessor'}` (model VideoMAEConfig)
- **videomt** -- `{'torchvision': 'VideomtVideoProcessor'}` (model VideomtConfig)
- **videoprism** -- `{'torchvision': 'LlavaOnevisionVideoProcessor'}` (model VideoPrismConfig)
- **vivit** -- `{'torchvision': 'VivitVideoProcessor'}` (model VivitConfig)
- **vjepa2** -- `{'torchvision': 'VJEPA2VideoProcessor'}` (model VJEPA2Config)

Menyampaikan `token=True` diperlukan ketika Anda ingin menggunakan model privat.

Contoh:

```python
>>> from transformers import AutoVideoProcessor

>>> # Download video processor from huggingface.co and cache.
>>> video_processor = AutoVideoProcessor.from_pretrained("llava-hf/llava-onevision-qwen2-0.5b-ov-hf")

>>> # If video processor files are in a directory (e.g. video processor was saved using *save_pretrained('./test/saved_model/')*)
>>> # video_processor = AutoVideoProcessor.from_pretrained("./test/saved_model/")
```

#### register[[transformers.AutoVideoProcessor.register]]

```python
register(config_class, video_processor_class: type | None = None, video_processor_classes: dict[str, type] | None = None, exist_ok: bool = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/video_processing_auto.py#L464)

**Parameter:**

config_class ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang sesuai dengan model yang akan didaftarkan.

video_processor_class ([BaseVideoProcessor](/docs/transformers/main/id/main_classes/video_processor#transformers.BaseVideoProcessor)) : Pemroses video yang akan didaftarkan.

video_processor_classes (`dict[str, type]`, *opsional*) : Kamus yang memetakan nama backend ke kelas video processor. Memungkinkan pendaftaran backend kustom. Contoh: `{"pil": MyPilVideoProcessor, "torchvision": MyTorchvisionVideoProcessor, "custom": MyCustomVideoProcessor}`

exist_ok (`bool`, *opsional*, defaultnya `False`) : Jika `True`, izinkan penulisan ulang pendaftaran yang sudah ada.

Daftarkan pemroses video baru untuk kelas ini.

## AutoProcessor[[transformers.AutoProcessor]]

#### transformers.AutoProcessor[[transformers.AutoProcessor]]

```python
transformers.AutoProcessor()
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/processing_auto.py#L127)

Ini adalah kelas pemroses generik yang akan diinstansiasi sebagai salah satu kelas pemroses dari pustaka saat dibuat dengan metode kelas [AutoProcessor.from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoProcessor.from_pretrained).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_pretrained[[transformers.AutoProcessor.from_pretrained]]

```python
from_pretrained(pretrained_model_name_or_path, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/processing_auto.py#L141)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Ini bisa berupa:  - string, *model id* dari feature_extractor pretrained yang dihosting di dalam model repo di huggingface.co. - path ke *directory* yang berisi file processor yang disimpan menggunakan metode `save_pretrained()`, misalnya `./my_model_directory/`.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat feature extractor model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, defaultnya `False`) : Apakah akan memaksa untuk (mengunduh ulang) file feature extractor dan menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.` Proxy digunakan pada setiap permintaan.

token (`str` atau *bool*, *opsional*) : Token yang akan digunakan sebagai otorisasi bearer HTTP untuk file jarak jauh. Jika `True`, akan menggunakan token yang dihasilkan saat menjalankan `hf auth login` (disimpan di `~/.huggingface`).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

return_unused_kwargs (`bool`, *opsional*, defaultnya `False`) : Jika `False`, fungsi ini hanya mengembalikan objek feature extractor akhir. Jika `True`, fungsi ini mengembalikan `Tuple(feature_extractor, unused_kwargs)` di mana *unused_kwargs* adalah kamus yang terdiri dari pasangan kunci/nilai yang kuncinya bukan merupakan atribut feature extractor: yaitu, bagian dari `kwargs` yang tidak digunakan untuk memperbarui `feature_extractor` dan diabaikan.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

kwargs (`dict[str, Any]`, *opsional*) : Nilai dalam kwargs untuk kunci apa pun yang merupakan atribut feature extractor akan digunakan untuk menimpa nilai yang dimuat. Perilaku terkait pasangan kunci/nilai yang kuncinya *bukan* atribut feature extractor dikendalikan oleh parameter kata kunci `return_unused_kwargs`.

Instansiasi salah satu kelas pemroses dari pustaka ini dari kosakata model pralatih.

Kelas prosesor yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan):

- **aimv2** -- [CLIPProcessor](/docs/transformers/main/id/model_doc/clip#transformers.CLIPProcessor) (model Aimv2Config)
- **align** -- [AlignProcessor](/docs/transformers/main/id/model_doc/align#transformers.AlignProcessor) (model AlignConfig)
- **altclip** -- [AltCLIPProcessor](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPProcessor) (model AltCLIPConfig)
- **aria** -- [AriaProcessor](/docs/transformers/main/id/model_doc/aria#transformers.AriaProcessor) (model AriaConfig)
- **audioflamingo3** -- [AudioFlamingo3Processor](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Processor) (model AudioFlamingo3Config)
- **aya_vision** -- [AyaVisionProcessor](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionProcessor) (model AyaVisionConfig)
- **bark** -- [BarkProcessor](/docs/transformers/main/id/model_doc/bark#transformers.BarkProcessor) (model BarkConfig)
- **blip** -- [BlipProcessor](/docs/transformers/main/id/model_doc/blip#transformers.BlipProcessor) (model BlipConfig)
- **blip-2** -- [Blip2Processor](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Processor) (model Blip2Config)
- **bridgetower** -- [BridgeTowerProcessor](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerProcessor) (model BridgeTowerConfig)
- **bros** -- [BrosProcessor](/docs/transformers/main/id/model_doc/bros#transformers.BrosProcessor) (model BrosConfig)
- **canary** -- [CanaryProcessor](/docs/transformers/main/id/model_doc/canary#transformers.CanaryProcessor) (model CanaryConfig)
- **chameleon** -- [ChameleonProcessor](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonProcessor) (model ChameleonConfig)
- **chinese_clip** -- [ChineseCLIPProcessor](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPProcessor) (model ChineseCLIPConfig)
- **clap** -- [ClapProcessor](/docs/transformers/main/id/model_doc/clap#transformers.ClapProcessor) (model ClapConfig)
- **clip** -- [CLIPProcessor](/docs/transformers/main/id/model_doc/clip#transformers.CLIPProcessor) (model CLIPConfig)
- **clipseg** -- [CLIPSegProcessor](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegProcessor) (model CLIPSegConfig)
- **clvp** -- [ClvpProcessor](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpProcessor) (model ClvpConfig)
- **cohere2_vision** -- [Cohere2VisionProcessor](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor) (model Cohere2VisionConfig)
- **cohere_asr** -- [CohereAsrProcessor](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrProcessor) (model CohereAsrConfig)
- **cohere_compass** -- [CohereCompassProcessor](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassProcessor) (model CohereCompassConfig)
- **colmodernvbert** -- [ColModernVBertProcessor](/docs/transformers/main/id/model_doc/colmodernvbert#transformers.ColModernVBertProcessor) (model ColModernVBertConfig)
- **colpali** -- [ColPaliProcessor](/docs/transformers/main/id/model_doc/colpali#transformers.ColPaliProcessor) (model ColPaliConfig)
- **colqwen2** -- [ColQwen2Processor](/docs/transformers/main/id/model_doc/colqwen2#transformers.ColQwen2Processor) (model ColQwen2Config)
- **cosmos3_edge** -- [Cosmos3EdgeProcessor](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeProcessor) (model Cosmos3EdgeConfig)
- **csm** -- [CsmProcessor](/docs/transformers/main/id/model_doc/csm#transformers.CsmProcessor) (model CsmConfig)
- **deepseek_ocr2** -- [DeepseekOcr2Processor](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Processor) (model DeepseekOcr2Config)
- **deepseek_vl** -- [DeepseekVLProcessor](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLProcessor) (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- [DeepseekVLHybridProcessor](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridProcessor) (model DeepseekVLHybridConfig)
- **dia** -- [DiaProcessor](/docs/transformers/main/id/model_doc/dia#transformers.DiaProcessor) (model DiaConfig)
- **diffusion_gemma** -- [Gemma4Processor](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Processor) (model DiffusionGemmaConfig)
- **edgetam** -- [Sam2Processor](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Processor) (model EdgeTamConfig)
- **embedding_gemma2** -- [EmbeddingGemma2Processor](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2Processor) (model EmbeddingGemma2Config)
- **emu3** -- [Emu3Processor](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Processor) (model Emu3Config)
- **ernie4_5_vl_moe** -- [Ernie4_5_VLMoeProcessor](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeProcessor) (model Ernie4_5_VLMoeConfig)
- **evolla** -- [EvollaProcessor](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaProcessor) (model EvollaConfig)
- **exaone4_5** -- [Exaone4_5_Processor](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Processor) (model Exaone4_5_Config)
- **flava** -- [FlavaProcessor](/docs/transformers/main/id/model_doc/flava#transformers.FlavaProcessor) (model FlavaConfig)
- **florence2** -- [Florence2Processor](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Processor) (model Florence2Config)
- **fun_asr_nano** -- [FunAsrNanoProcessor](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoProcessor) (model FunAsrNanoConfig)
- **fuyu** -- [FuyuProcessor](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuProcessor) (model FuyuConfig)
- **gemma3** -- [Gemma3Processor](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Processor) (model Gemma3Config)
- **gemma3n** -- [Gemma3nProcessor](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nProcessor) (model Gemma3nConfig)
- **gemma4** -- [Gemma4Processor](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Processor) (model Gemma4Config)
- **gemma4_unified** -- [Gemma4UnifiedProcessor](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedProcessor) (model Gemma4UnifiedConfig)
- **git** -- [GitProcessor](/docs/transformers/main/id/model_doc/git#transformers.GitProcessor) (model GitConfig)
- **glm46v** -- [Glm46VProcessor](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VProcessor) (model Glm46VConfig)
- **glm4v** -- [Glm4vProcessor](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vProcessor) (model Glm4vConfig)
- **glm4v_moe** -- [Glm4vProcessor](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vProcessor) (model Glm4vMoeConfig)
- **glm5_next** -- [Glm5NextProcessor](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextProcessor) (model Glm5NextConfig)
- **glm_image** -- [GlmImageProcessor](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageProcessor) (model GlmImageConfig)
- **glmasr** -- [GlmAsrProcessor](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrProcessor) (model GlmAsrConfig)
- **got_ocr2** -- [GotOcr2Processor](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Processor) (model GotOcr2Config)
- **gradium_phonon_tts** -- [PocketTTSProcessor](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSProcessor) (model GradiumPhononTTSConfig)
- **granite4_vision** -- [Granite4VisionProcessor](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionProcessor) (model Granite4VisionConfig)
- **granite_speech** -- [GraniteSpeechProcessor](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechProcessor) (model GraniteSpeechConfig)
- **granite_speech5_ctc** -- [GraniteSpeech5Processor](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5Processor) (model GraniteSpeech5CTCConfig)
- **granite_speech_plus** -- [GraniteSpeechProcessor](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechProcessor) (model GraniteSpeechPlusConfig)
- **grounding-dino** -- [GroundingDinoProcessor](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoProcessor) (model GroundingDinoConfig)
- **groupvit** -- [CLIPProcessor](/docs/transformers/main/id/model_doc/clip#transformers.CLIPProcessor) (model GroupViTConfig)
- **higgs_audio_v2** -- [HiggsAudioV2Processor](/docs/transformers/main/id/model_doc/higgs_audio_v2#transformers.HiggsAudioV2Processor) (model HiggsAudioV2Config)
- **hubert** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model HubertConfig)
- **hunyuan_vl** -- [HunYuanVLProcessor](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLProcessor) (model HunYuanVLConfig)
- **hyperclovax_vision_v2** -- [HyperCLOVAXVisionV2Processor](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Processor) (model HyperCLOVAXVisionV2Config)
- **idefics** -- [IdeficsProcessor](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsProcessor) (model IdeficsConfig)
- **idefics2** -- [Idefics2Processor](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Processor) (model Idefics2Config)
- **idefics3** -- [Idefics3Processor](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Processor) (model Idefics3Config)
- **inkling_mm_model** -- [InklingProcessor](/docs/transformers/main/id/model_doc/inkling#transformers.InklingProcessor) (model InklingConfig)
- **instructblip** -- [InstructBlipProcessor](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipProcessor) (model InstructBlipConfig)
- **instructblipvideo** -- [InstructBlipVideoProcessor](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoProcessor) (model InstructBlipVideoConfig)
- **internvl** -- [InternVLProcessor](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLProcessor) (model InternVLConfig)
- **janus** -- [JanusProcessor](/docs/transformers/main/id/model_doc/janus#transformers.JanusProcessor) (model JanusConfig)
- **kimi_k25** -- [Kimi_K25Processor](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Processor) (model Kimi_K25Config)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextProcessor](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextProcessor) (model KyutaiSpeechToTextConfig)
- **lasr_ctc** -- [LasrProcessor](/docs/transformers/main/id/model_doc/lasr#transformers.LasrProcessor) (model LasrCTCConfig)
- **lasr_encoder** -- [LasrProcessor](/docs/transformers/main/id/model_doc/lasr#transformers.LasrProcessor) (model LasrEncoderConfig)
- **layoutlmv2** -- [LayoutLMv2Processor](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Processor) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3Processor](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Processor) (model LayoutLMv3Config)
- **layoutxlm** -- [LayoutXLMProcessor](/docs/transformers/main/id/model_doc/layoutxlm#transformers.LayoutXLMProcessor) (model LayoutXLMConfig)
- **lfm2_vl** -- [Lfm2VlProcessor](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlProcessor) (model Lfm2VlConfig)
- **lighton_ocr** -- [LightOnOcrProcessor](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrProcessor) (model LightOnOcrConfig)
- **llama4** -- [Llama4Processor](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Processor) (model Llama4Config)
- **llava** -- [LlavaProcessor](/docs/transformers/main/id/model_doc/llava#transformers.LlavaProcessor) (model LlavaConfig)
- **llava_next** -- [LlavaNextProcessor](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextProcessor) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoProcessor](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoProcessor) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionProcessor](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionProcessor) (model LlavaOnevisionConfig)
- **markuplm** -- [MarkupLMProcessor](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMProcessor) (model MarkupLMConfig)
- **metaclip_2** -- [CLIPProcessor](/docs/transformers/main/id/model_doc/clip#transformers.CLIPProcessor) (model MetaClip2Config)
- **mgp-str** -- [MgpstrProcessor](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrProcessor) (model MgpstrConfig)
- **minicpm_ocr** -- [MiniCPMOcrProcessor](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrProcessor) (model MiniCPMOcrConfig)
- **minicpmv4_6** -- [MiniCPMV4_6Processor](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Processor) (model MiniCPMV4_6Config)
- **minicpmv4_7** -- [MiniCPMV4_7Processor](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Processor) (model MiniCPMV4_7Config)
- **minimax_m3_vl** -- [MiniMaxM3VLProcessor](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLProcessor) (model MiniMaxM3VLConfig)
- **mistral3** -- [PixtralProcessor](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralProcessor) (model Mistral3Config)
- **mistral_large4** -- [PixtralProcessor](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralProcessor) (model MistralLarge4Config)
- **mllama** -- [MllamaProcessor](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaProcessor) (model MllamaConfig)
- **mm-grounding-dino** -- [GroundingDinoProcessor](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoProcessor) (model MMGroundingDinoConfig)
- **modernvbert** -- [Idefics3Processor](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Processor) (model ModernVBertConfig)
- **molmo2** -- [Molmo2Processor](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Processor) (model Molmo2Config)
- **moonshine** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model MoonshineConfig)
- **moonshine_streaming** -- [MoonshineStreamingProcessor](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingProcessor) (model MoonshineStreamingConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeProcessor](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeProcessor) (model MossTranscribeDiarizeConfig)
- **muse_glimmer** -- [MuseGlimmerProcessor](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerProcessor) (model MuseGlimmerConfig)
- **muse_spark** -- [MuseSparkProcessor](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkProcessor) (model MuseSparkConfig)
- **musicflamingo** -- [MusicFlamingoProcessor](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoProcessor) (model MusicFlamingoConfig)
- **musicgen** -- [MusicgenProcessor](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenProcessor) (model MusicgenConfig)
- **musicgen_melody** -- [MusicgenMelodyProcessor](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyProcessor) (model MusicgenMelodyConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeProcessor](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeProcessor) (model Nemotron3_5TranscribeConfig)
- **nemotron3_5_asr** -- [Nemotron3_5AsrProcessor](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrProcessor) (model Nemotron3_5AsrConfig)
- **nemotron3_diarization** -- [Nemotron3DiarizationProcessor](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationProcessor) (model Nemotron3DiarizationConfig)
- **nemotron_asr_streaming** -- [NemotronAsrStreamingProcessor](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingProcessor) (model NemotronAsrStreamingConfig)
- **nemotron_h_omni** -- [NemotronH_Omni_Reasoning_V3Processor](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3Processor) (model NemotronH_Omni_Reasoning_V3_Config)
- **neomme** -- [NeoMMEProcessor](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEProcessor) (model NeoMMEConfig)
- **nougat** -- [NougatProcessor](/docs/transformers/main/id/model_doc/nougat#transformers.NougatProcessor) (model NougatConfig)
- **omdet-turbo** -- [OmDetTurboProcessor](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboProcessor) (model OmDetTurboConfig)
- **oneformer** -- [OneFormerProcessor](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerProcessor) (model OneFormerConfig)
- **ovis2** -- [Ovis2Processor](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Processor) (model Ovis2Config)
- **owlv2** -- [Owlv2Processor](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Processor) (model Owlv2Config)
- **owlvit** -- [OwlViTProcessor](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTProcessor) (model OwlViTConfig)
- **paddleocr_vl** -- [PaddleOCRVLProcessor](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLProcessor) (model PaddleOCRVLConfig)
- **paligemma** -- [PaliGemmaProcessor](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaProcessor) (model PaliGemmaConfig)
- **parakeet_ctc** -- [ParakeetProcessor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetProcessor) (model ParakeetCTCConfig)
- **parakeet_rnnt** -- [ParakeetProcessor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetProcessor) (model ParakeetRNNTConfig)
- **parakeet_tdt** -- [ParakeetProcessor](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetProcessor) (model ParakeetTDTConfig)
- **pe_audio** -- [PeAudioProcessor](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioProcessor) (model PeAudioConfig)
- **pe_audio_video** -- [PeAudioVideoProcessor](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoProcessor) (model PeAudioVideoConfig)
- **pe_video** -- [PeVideoProcessor](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoProcessor) (model PeVideoConfig)
- **perception_lm** -- [PerceptionLMProcessor](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMProcessor) (model PerceptionLMConfig)
- **phi4_multimodal** -- [Phi4MultimodalProcessor](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalProcessor) (model Phi4MultimodalConfig)
- **pi0** -- [PI0Processor](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Processor) (model PI0Config)
- **pix2struct** -- [Pix2StructProcessor](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructProcessor) (model Pix2StructConfig)
- **pixtral** -- [PixtralProcessor](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralProcessor) (model PixtralVisionConfig)
- **pocket_tts** -- [PocketTTSProcessor](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSProcessor) (model PocketTTSConfig)
- **pop2piano** -- [Pop2PianoProcessor](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoProcessor) (model Pop2PianoConfig)
- **pp_chart2table** -- [PPChart2TableProcessor](/docs/transformers/main/id/model_doc/pp_chart2table#transformers.PPChart2TableProcessor) (model PPChart2TableConfig)
- **pp_formulanet** -- [PPFormulaNetProcessor](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetProcessor) (model PPFormulaNetConfig)
- **qianfan_ocr** -- [QianfanOCRProcessor](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRProcessor) (model QianfanOCRConfig)
- **qwen2_5_omni** -- [Qwen2_5OmniProcessor](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniProcessor) (model Qwen2_5OmniConfig)
- **qwen2_5_vl** -- [Qwen2_5_VLProcessor](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLProcessor) (model Qwen2_5_VLConfig)
- **qwen2_audio** -- [Qwen2AudioProcessor](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioProcessor) (model Qwen2AudioConfig)
- **qwen2_vl** -- [Qwen2VLProcessor](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLProcessor) (model Qwen2VLConfig)
- **qwen3_5** -- [Qwen3VLProcessor](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLProcessor) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3VLProcessor](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLProcessor) (model Qwen3_5MoeConfig)
- **qwen3_asr** -- [Qwen3ASRProcessor](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRProcessor) (model Qwen3ASRConfig)
- **qwen3_omni_moe** -- [Qwen3OmniMoeProcessor](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeProcessor) (model Qwen3OmniMoeConfig)
- **qwen3_vl** -- [Qwen3VLProcessor](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLProcessor) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen3VLProcessor](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLProcessor) (model Qwen3VLMoeConfig)
- **qwen4_exp** -- [Qwen3VLProcessor](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLProcessor) (model Qwen4ExpConfig)
- **sam** -- [SamProcessor](/docs/transformers/main/id/model_doc/sam#transformers.SamProcessor) (model SamConfig)
- **sam2** -- [Sam2Processor](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Processor) (model Sam2Config)
- **sam2_video** -- [Sam2VideoProcessor](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoProcessor) (model Sam2VideoConfig)
- **sam3** -- [Sam3Processor](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Processor) (model Sam3Config)
- **sam3_lite_text** -- [Sam3Processor](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Processor) (model Sam3LiteTextConfig)
- **sam3_tracker** -- [Sam3TrackerProcessor](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerProcessor) (model Sam3TrackerConfig)
- **sam3_tracker_video** -- [Sam3TrackerVideoProcessor](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoProcessor) (model Sam3TrackerVideoConfig)
- **sam3_video** -- [Sam3VideoProcessor](/docs/transformers/main/id/model_doc/sam3_video#transformers.Sam3VideoProcessor) (model Sam3VideoConfig)
- **sam_hq** -- [SamHQProcessor](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQProcessor) (model SamHQConfig)
- **seamless_m4t** -- [SeamlessM4TProcessor](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TProcessor) (model SeamlessM4TConfig)
- **sew** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model SEWConfig)
- **sew-d** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model SEWDConfig)
- **siglip** -- [SiglipProcessor](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipProcessor) (model SiglipConfig)
- **siglip2** -- [Siglip2Processor](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Processor) (model Siglip2Config)
- **smolvlm** -- [SmolVLMProcessor](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMProcessor) (model SmolVLMConfig)
- **speech_to_text** -- [Speech2TextProcessor](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextProcessor) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5Processor](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Processor) (model SpeechT5Config)
- **step3p7** -- [Step3p7Processor](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Processor) (model Step3p7Config)
- **t5gemma2** -- [Gemma3Processor](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Processor) (model T5Gemma2Config)
- **t5gemma2_encoder** -- [Gemma3Processor](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Processor) (model T5Gemma2EncoderConfig)
- **tipsv2** -- [Tipsv2Processor](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Processor) (model Tipsv2Config)
- **trocr** -- [TrOCRProcessor](/docs/transformers/main/id/model_doc/trocr#transformers.TrOCRProcessor) (model TrOCRConfig)
- **tvp** -- [TvpProcessor](/docs/transformers/main/id/model_doc/tvp#transformers.TvpProcessor) (model TvpConfig)
- **udop** -- [UdopProcessor](/docs/transformers/main/id/model_doc/udop#transformers.UdopProcessor) (model UdopConfig)
- **ultravbert** -- [Idefics3Processor](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Processor) (model UltraVBertConfig)
- **unispeech** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model UniSpeechConfig)
- **unispeech-sat** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model UniSpeechSatConfig)
- **unlimited_ocr** -- [UnlimitedOcrProcessor](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrProcessor) (model UnlimitedOcrConfig)
- **vibevoice** -- [VibeVoiceProcessor](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceProcessor) (model VibeVoiceConfig)
- **vibevoice_asr** -- [VibeVoiceAsrProcessor](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrProcessor) (model VibeVoiceAsrConfig)
- **video_llama_3** -- [VideoLlama3Processor](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Processor) (model VideoLlama3Config)
- **video_llava** -- [VideoLlavaProcessor](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaProcessor) (model VideoLlavaConfig)
- **videoprism** -- [VideoPrismProcessor](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismProcessor) (model VideoPrismConfig)
- **vilt** -- [ViltProcessor](/docs/transformers/main/id/model_doc/vilt#transformers.ViltProcessor) (model ViltConfig)
- **vipllava** -- [LlavaProcessor](/docs/transformers/main/id/model_doc/llava#transformers.LlavaProcessor) (model VipLlavaConfig)
- **vision-text-dual-encoder** -- [VisionTextDualEncoderProcessor](/docs/transformers/main/id/model_doc/vision-text-dual-encoder#transformers.VisionTextDualEncoderProcessor) (model VisionTextDualEncoderConfig)
- **voxtral** -- [VoxtralProcessor](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralProcessor) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeProcessor](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeProcessor) (model VoxtralRealtimeConfig)
- **wav2vec2** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertProcessor](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertProcessor) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [Wav2Vec2Processor](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Processor) (model WavLMConfig)
- **whisper** -- [WhisperProcessor](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperProcessor) (model WhisperConfig)

Menyampaikan `token=True` diperlukan ketika Anda ingin menggunakan model privat.

Contoh:

```python
>>> from transformers import AutoProcessor

>>> # Download processor from huggingface.co and cache.
>>> processor = AutoProcessor.from_pretrained("facebook/wav2vec2-base-960h")

>>> # If processor files are in a directory (e.g. processor was saved using *save_pretrained('./test/saved_model/')*)
>>> # processor = AutoProcessor.from_pretrained("./test/saved_model/")
```

#### register[[transformers.AutoProcessor.register]]

```python
register(config_class, processor_class, exist_ok = False)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/processing_auto.py#L373)

**Parameter:**

config_class ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Konfigurasi yang sesuai dengan model yang akan didaftarkan.

processor_class ([ProcessorMixin](/docs/transformers/main/id/main_classes/processors#transformers.ProcessorMixin)) : Prosesor yang akan didaftarkan.

Daftarkan prosesor baru untuk kelas ini.

## Kelas model generik

Kelas auto berikut tersedia untuk menginstansiasi kelas model dasar tanpa head spesifik.

### AutoModel[[transformers.AutoModel]]

#### transformers.AutoModel[[transformers.AutoModel]]

```python
transformers.AutoModel(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2297)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dasar dari pustaka saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModel.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [ASTConfig](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTConfig) configuration class: [ASTModel](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTModel) (ASTConfig model) - [AXK1Config](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Config) configuration class: [AXK1Model](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Model) (AXK1Config model) - [AXK2Config](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Config) configuration class: [AXK2Model](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Model) (AXK2Config model) - [AfmoeConfig](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeConfig) configuration class: [AfmoeModel](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeModel) (AfmoeConfig model) - [Aimv2Config](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2Config) configuration class: [Aimv2Model](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2Model) (Aimv2Config model) - [Aimv2VisionConfig](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2VisionConfig) configuration class: [Aimv2VisionModel](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2VisionModel) (Aimv2VisionConfig model) - [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig) configuration class: [AlbertModel](/docs/transformers/main/id/model_doc/albert#transformers.AlbertModel) (AlbertConfig model) - [AlignConfig](/docs/transformers/main/id/model_doc/align#transformers.AlignConfig) configuration class: [AlignModel](/docs/transformers/main/id/model_doc/align#transformers.AlignModel) (AlignConfig model) - [AltCLIPConfig](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPConfig) configuration class: [AltCLIPModel](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPModel) (AltCLIPConfig model) - [ApertusConfig](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusConfig) configuration class: [ApertusModel](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusModel) (ApertusConfig model) - [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig) configuration class: [ArceeModel](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeModel) (ArceeConfig model) - [AriaConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaConfig) configuration class: [AriaModel](/docs/transformers/main/id/model_doc/aria#transformers.AriaModel) (AriaConfig model) - [AriaTextConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextConfig) configuration class: [AriaTextModel](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextModel) (AriaTextConfig model) - [AudioFlamingo3Config](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Config) configuration class: [AudioFlamingo3Model](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Model) (AudioFlamingo3Config model) - [AudioFlamingo3EncoderConfig](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3EncoderConfig) configuration class: [AudioFlamingo3Encoder](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Encoder) (AudioFlamingo3EncoderConfig model) - [AutoformerConfig](/docs/transformers/main/id/model_doc/autoformer#transformers.AutoformerConfig) configuration class: [AutoformerModel](/docs/transformers/main/id/model_doc/autoformer#transformers.AutoformerModel) (AutoformerConfig model) - [AyaVisionConfig](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionConfig) configuration class: [AyaVisionModel](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionModel) (AyaVisionConfig model) - [BailingHybridConfig](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridConfig) configuration class: [BailingHybridModel](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridModel) (BailingHybridConfig model) - [BambaConfig](/docs/transformers/main/id/model_doc/bamba#transformers.BambaConfig) configuration class: [BambaModel](/docs/transformers/main/id/model_doc/bamba#transformers.BambaModel) (BambaConfig model) - [BarkConfig](/docs/transformers/main/id/model_doc/bark#transformers.BarkConfig) configuration class: [BarkModel](/docs/transformers/main/id/model_doc/bark#transformers.BarkModel) (BarkConfig model) - [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig) configuration class: [BartModel](/docs/transformers/main/id/model_doc/bart#transformers.BartModel) (BartConfig model) - [BeamConfig](/docs/transformers/main/id/model_doc/beam#transformers.BeamConfig) configuration class: [BeamModel](/docs/transformers/main/id/model_doc/beam#transformers.BeamModel) (BeamConfig model) - [BeitConfig](/docs/transformers/main/id/model_doc/beit#transformers.BeitConfig) configuration class: [BeitModel](/docs/transformers/main/id/model_doc/beit#transformers.BeitModel) (BeitConfig model) - [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) configuration class: [BertModel](/docs/transformers/main/id/model_doc/bert#transformers.BertModel) (BertConfig model) - [BertGenerationConfig](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationConfig) configuration class: [BertGenerationEncoder](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationEncoder) (BertGenerationConfig model) - [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) configuration class: [BigBirdModel](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdModel) (BigBirdConfig model) - [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig) configuration class: [BigBirdPegasusModel](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusModel) (BigBirdPegasusConfig model) - [BioGptConfig](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptConfig) configuration class: [BioGptModel](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptModel) (BioGptConfig model) - [BitConfig](/docs/transformers/main/id/model_doc/bit#transformers.BitConfig) configuration class: [BitModel](/docs/transformers/main/id/model_doc/bit#transformers.BitModel) (BitConfig model) - [BitNetConfig](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetConfig) configuration class: [BitNetModel](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetModel) (BitNetConfig model) - [BlenderbotConfig](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotConfig) configuration class: [BlenderbotModel](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotModel) (BlenderbotConfig model) - [BlenderbotSmallConfig](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallConfig) configuration class: [BlenderbotSmallModel](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallModel) (BlenderbotSmallConfig model) - [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config) configuration class: [Blip2Model](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Model) (Blip2Config model) - [Blip2QFormerConfig](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2QFormerConfig) configuration class: [Blip2QFormerModel](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2QFormerModel) (Blip2QFormerConfig model) - [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig) configuration class: [BlipModel](/docs/transformers/main/id/model_doc/blip#transformers.BlipModel) (BlipConfig model) - [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) configuration class: [BloomModel](/docs/transformers/main/id/model_doc/bloom#transformers.BloomModel) (BloomConfig model) - [BltConfig](/docs/transformers/main/id/model_doc/blt#transformers.BltConfig) configuration class: [BltModel](/docs/transformers/main/id/model_doc/blt#transformers.BltModel) (BltConfig model) - [BridgeTowerConfig](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerConfig) configuration class: [BridgeTowerModel](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerModel) (BridgeTowerConfig model) - [BrosConfig](/docs/transformers/main/id/model_doc/bros#transformers.BrosConfig) configuration class: [BrosModel](/docs/transformers/main/id/model_doc/bros#transformers.BrosModel) (BrosConfig model) - [CLIPConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPConfig) configuration class: [CLIPModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPModel) (CLIPConfig model) - [CLIPSegConfig](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegConfig) configuration class: [CLIPSegModel](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegModel) (CLIPSegConfig model) - [CLIPTextConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTextConfig) configuration class: [CLIPTextModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTextModel) (CLIPTextConfig model) - [CLIPVisionConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPVisionConfig) configuration class: [CLIPVisionModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPVisionModel) (CLIPVisionConfig model) - [CTRLConfig](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLConfig) configuration class: [CTRLModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLModel) (CTRLConfig model) - [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) configuration class: [CamembertModel](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertModel) (CamembertConfig model) - [CanaryConfig](/docs/transformers/main/id/model_doc/canary#transformers.CanaryConfig) configuration class: [CanaryModel](/docs/transformers/main/id/model_doc/canary#transformers.CanaryModel) (CanaryConfig model) - [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig) configuration class: [CanineModel](/docs/transformers/main/id/model_doc/canine#transformers.CanineModel) (CanineConfig model) - [ChameleonConfig](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonConfig) configuration class: [ChameleonModel](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonModel) (ChameleonConfig model) - [ChineseCLIPConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPConfig) configuration class: [ChineseCLIPModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPModel) (ChineseCLIPConfig model) - [ChineseCLIPVisionConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPVisionConfig) configuration class: [ChineseCLIPVisionModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPVisionModel) (ChineseCLIPVisionConfig model) - [ClapConfig](/docs/transformers/main/id/model_doc/clap#transformers.ClapConfig) configuration class: [ClapModel](/docs/transformers/main/id/model_doc/clap#transformers.ClapModel) (ClapConfig model) - [ClvpConfig](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpConfig) configuration class: [ClvpModelForConditionalGeneration](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpModelForConditionalGeneration) (ClvpConfig model) - [CodeGenConfig](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenConfig) configuration class: [CodeGenModel](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenModel) (CodeGenConfig model) - [Cohere2Config](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2Config) configuration class: [Cohere2Model](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2Model) (Cohere2Config model) - [Cohere2MoeConfig](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeConfig) configuration class: [Cohere2MoeModel](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeModel) (Cohere2MoeConfig model) - [Cohere2VisionConfig](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionConfig) configuration class: [Cohere2VisionModel](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionModel) (Cohere2VisionConfig model) - [CohereAsrConfig](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrConfig) configuration class: [CohereAsrModel](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrModel) (CohereAsrConfig model) - [CohereCompassConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassConfig) configuration class: [CohereCompassModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassModel) (CohereCompassConfig model) - [CohereCompassTextConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextConfig) configuration class: [CohereCompassTextModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextModel) (CohereCompassTextConfig model) - [CohereCompassVisionConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassVisionConfig) configuration class: [CohereCompassVisionModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassVisionModel) (CohereCompassVisionConfig model) - [CohereConfig](/docs/transformers/main/id/model_doc/cohere#transformers.CohereConfig) configuration class: [CohereModel](/docs/transformers/main/id/model_doc/cohere#transformers.CohereModel) (CohereConfig model) - [ConditionalDetrConfig](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrConfig) configuration class: [ConditionalDetrModel](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrModel) (ConditionalDetrConfig model) - [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig) configuration class: [ConvBertModel](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertModel) (ConvBertConfig model) - [ConvNextConfig](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextConfig) configuration class: [ConvNextModel](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextModel) (ConvNextConfig model) - [ConvNextV2Config](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2Config) configuration class: [ConvNextV2Model](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2Model) (ConvNextV2Config model) - [Cosmos3EdgeConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeConfig) configuration class: [Cosmos3EdgeModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeModel) (Cosmos3EdgeConfig model) - [Cosmos3EdgeTextConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeTextConfig) configuration class: [Cosmos3EdgeTextModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeTextModel) (Cosmos3EdgeTextConfig model) - [Cosmos3EdgeVisionConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeVisionConfig) configuration class: [Cosmos3EdgeVisionModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeVisionModel) (Cosmos3EdgeVisionConfig model) - [Cosmos3OmniConfig](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniConfig) configuration class: [Cosmos3OmniModel](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniModel) (Cosmos3OmniConfig model) - [CpmAntConfig](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntConfig) configuration class: [CpmAntModel](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntModel) (CpmAntConfig model) - [CsmConfig](/docs/transformers/main/id/model_doc/csm#transformers.CsmConfig) configuration class: [CsmForConditionalGeneration](/docs/transformers/main/id/model_doc/csm#transformers.CsmForConditionalGeneration) (CsmConfig model) - [CvtConfig](/docs/transformers/main/id/model_doc/cvt#transformers.CvtConfig) configuration class: [CvtModel](/docs/transformers/main/id/model_doc/cvt#transformers.CvtModel) (CvtConfig model) - [CwmConfig](/docs/transformers/main/id/model_doc/cwm#transformers.CwmConfig) configuration class: [CwmModel](/docs/transformers/main/id/model_doc/cwm#transformers.CwmModel) (CwmConfig model) - [DFineConfig](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineConfig) configuration class: [DFineModel](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineModel) (DFineConfig model) - [DINOv3ConvNextConfig](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ConvNextConfig) configuration class: [DINOv3ConvNextModel](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ConvNextModel) (DINOv3ConvNextConfig model) - [DINOv3ViTConfig](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ViTConfig) configuration class: [DINOv3ViTModel](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ViTModel) (DINOv3ViTConfig model) - [DPRConfig](/docs/transformers/main/id/model_doc/dpr#transformers.DPRConfig) configuration class: [DPRQuestionEncoder](/docs/transformers/main/id/model_doc/dpr#transformers.DPRQuestionEncoder) (DPRConfig model) - [DPTConfig](/docs/transformers/main/id/model_doc/dpt#transformers.DPTConfig) configuration class: [DPTModel](/docs/transformers/main/id/model_doc/dpt#transformers.DPTModel) (DPTConfig model) - [DabDetrConfig](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrConfig) configuration class: [DabDetrModel](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrModel) (DabDetrConfig model) - [DacConfig](/docs/transformers/main/id/model_doc/dac#transformers.DacConfig) configuration class: [DacModel](/docs/transformers/main/id/model_doc/dac#transformers.DacModel) (DacConfig model) - [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig) configuration class: [Data2VecAudioModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioModel) (Data2VecAudioConfig model) - [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) configuration class: [Data2VecTextModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextModel) (Data2VecTextConfig model) - [Data2VecVisionConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionConfig) configuration class: [Data2VecVisionModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionModel) (Data2VecVisionConfig model) - [DbrxConfig](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxConfig) configuration class: [DbrxModel](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxModel) (DbrxConfig model) - [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig) configuration class: [DebertaModel](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaModel) (DebertaConfig model) - [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config) configuration class: [DebertaV2Model](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Model) (DebertaV2Config model) - [DecisionTransformerConfig](/docs/transformers/main/id/model_doc/decision_transformer#transformers.DecisionTransformerConfig) configuration class: [DecisionTransformerModel](/docs/transformers/main/id/model_doc/decision_transformer#transformers.DecisionTransformerModel) (DecisionTransformerConfig model) - [DeepseekOcr2Config](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Config) configuration class: [DeepseekOcr2Model](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Model) (DeepseekOcr2Config model) - [DeepseekV2Config](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Config) configuration class: [DeepseekV2Model](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Model) (DeepseekV2Config model) - [DeepseekV32Config](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32Config) configuration class: [DeepseekV32Model](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32Model) (DeepseekV32Config model) - [DeepseekV3Config](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Config) configuration class: [DeepseekV3Model](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Model) (DeepseekV3Config model) - [DeepseekV4Config](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4Config) configuration class: [DeepseekV4Model](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4Model) (DeepseekV4Config model) - [DeepseekVLConfig](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLConfig) configuration class: [DeepseekVLModel](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLModel) (DeepseekVLConfig model) - [DeepseekVLHybridConfig](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridConfig) configuration class: [DeepseekVLHybridModel](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridModel) (DeepseekVLHybridConfig model) - [DeformableDetrConfig](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrConfig) configuration class: [DeformableDetrModel](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrModel) (DeformableDetrConfig model) - [DeiTConfig](/docs/transformers/main/id/model_doc/deit#transformers.DeiTConfig) configuration class: [DeiTModel](/docs/transformers/main/id/model_doc/deit#transformers.DeiTModel) (DeiTConfig model) - [Deimv2Config](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2Config) configuration class: [Deimv2Model](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2Model) (Deimv2Config model) - [DepthProConfig](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProConfig) configuration class: [DepthProModel](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProModel) (DepthProConfig model) - [DetrConfig](/docs/transformers/main/id/model_doc/detr#transformers.DetrConfig) configuration class: [DetrModel](/docs/transformers/main/id/model_doc/detr#transformers.DetrModel) (DetrConfig model) - [DiaConfig](/docs/transformers/main/id/model_doc/dia#transformers.DiaConfig) configuration class: [DiaModel](/docs/transformers/main/id/model_doc/dia#transformers.DiaModel) (DiaConfig model) - [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig) configuration class: [DiffLlamaModel](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaModel) (DiffLlamaConfig model) - [DiffusionGemmaConfig](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaConfig) configuration class: [DiffusionGemmaModel](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaModel) (DiffusionGemmaConfig model) - [DinatConfig](/docs/transformers/main/id/model_doc/dinat#transformers.DinatConfig) configuration class: [DinatModel](/docs/transformers/main/id/model_doc/dinat#transformers.DinatModel) (DinatConfig model) - [Dinov2Config](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2Config) configuration class: [Dinov2Model](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2Model) (Dinov2Config model) - [Dinov2WithRegistersConfig](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersConfig) configuration class: [Dinov2WithRegistersModel](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersModel) (Dinov2WithRegistersConfig model) - [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig) configuration class: [DistilBertModel](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertModel) (DistilBertConfig model) - [DogeConfig](/docs/transformers/main/id/model_doc/doge#transformers.DogeConfig) configuration class: [DogeModel](/docs/transformers/main/id/model_doc/doge#transformers.DogeModel) (DogeConfig model) - [DonutSwinConfig](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinConfig) configuration class: [DonutSwinModel](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinModel) (DonutSwinConfig model) - [Dots1Config](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1Config) configuration class: [Dots1Model](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1Model) (Dots1Config model) - [EdgeTamConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamConfig) configuration class: [EdgeTamModel](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamModel) (EdgeTamConfig model) - [EdgeTamVideoConfig](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoConfig) configuration class: [EdgeTamVideoModel](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoModel) (EdgeTamVideoConfig model) - [EdgeTamVisionConfig](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamVisionConfig) configuration class: [EdgeTamVisionModel](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamVisionModel) (EdgeTamVisionConfig model) - [EfficientLoFTRConfig](/docs/transformers/main/id/model_doc/efficientloftr#transformers.EfficientLoFTRConfig) configuration class: [EfficientLoFTRModel](/docs/transformers/main/id/model_doc/efficientloftr#transformers.EfficientLoFTRModel) (EfficientLoFTRConfig model) - [EfficientNetConfig](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetConfig) configuration class: [EfficientNetModel](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetModel) (EfficientNetConfig model) - [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) configuration class: [ElectraModel](/docs/transformers/main/id/model_doc/electra#transformers.ElectraModel) (ElectraConfig model) - [EmbeddingGemma2Config](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2Config) configuration class: [EmbeddingGemma2Model](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2Model) (EmbeddingGemma2Config model) - [EmbeddingGemma2TextConfig](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2TextConfig) configuration class: [EmbeddingGemma2TextModel](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2TextModel) (EmbeddingGemma2TextConfig model) - [Emu3Config](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Config) configuration class: [Emu3Model](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Model) (Emu3Config model) - [EncodecConfig](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecConfig) configuration class: [EncodecModel](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecModel) (EncodecConfig model) - [Ernie4_5Config](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5Config) configuration class: [Ernie4_5Model](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5Model) (Ernie4_5Config model) - [Ernie4_5_MoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeConfig) configuration class: [Ernie4_5_MoeModel](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeModel) (Ernie4_5_MoeConfig model) - [Ernie4_5_VLMoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeConfig) configuration class: [Ernie4_5_VLMoeModel](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeModel) (Ernie4_5_VLMoeConfig model) - [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) configuration class: [ErnieModel](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieModel) (ErnieConfig model) - [EsmConfig](/docs/transformers/main/id/model_doc/esm#transformers.EsmConfig) configuration class: [EsmModel](/docs/transformers/main/id/model_doc/esm#transformers.EsmModel) (EsmConfig model) - [EsmcConfig](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcConfig) configuration class: [EsmcModel](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcModel) (EsmcConfig model) - [EuroBertConfig](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertConfig) configuration class: [EuroBertModel](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertModel) (EuroBertConfig model) - [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) configuration class: [EvollaModel](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaModel) (EvollaConfig model) - [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) configuration class: [Exaone4Model](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Model) (Exaone4Config model) - [Exaone4_5_Config](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Config) configuration class: [Exaone4_5_Model](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Model) (Exaone4_5_Config model) - [Exaone4_5_VisionConfig](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_VisionConfig) configuration class: [Exaone4_5_VisionModel](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_VisionModel) (Exaone4_5_VisionConfig model) - [ExaoneMoeConfig](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeConfig) configuration class: [ExaoneMoeModel](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeModel) (ExaoneMoeConfig model) - [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig) configuration class: [FNetModel](/docs/transformers/main/id/model_doc/fnet#transformers.FNetModel) (FNetConfig model) - [FSMTConfig](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTConfig) configuration class: [FSMTModel](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTModel) (FSMTConfig model) - [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig) configuration class: [FalconModel](/docs/transformers/main/id/model_doc/falcon#transformers.FalconModel) (FalconConfig model) - [FalconH1Config](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1Config) configuration class: [FalconH1Model](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1Model) (FalconH1Config model) - [FalconMambaConfig](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaConfig) configuration class: [FalconMambaModel](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaModel) (FalconMambaConfig model) - [FastSpeech2ConformerConfig](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerConfig) configuration class: [FastSpeech2ConformerModel](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerModel) (FastSpeech2ConformerConfig model) - [FastSpeech2ConformerWithHifiGanConfig](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerWithHifiGanConfig) configuration class: [FastSpeech2ConformerWithHifiGan](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerWithHifiGan) (FastSpeech2ConformerWithHifiGanConfig model) - [FastVlmConfig](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmConfig) configuration class: [FastVlmModel](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmModel) (FastVlmConfig model) - [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig) configuration class: [FlaubertModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertModel) (FlaubertConfig model) - [FlavaConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaConfig) configuration class: [FlavaModel](/docs/transformers/main/id/model_doc/flava#transformers.FlavaModel) (FlavaConfig model) - [FlexOlmoConfig](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoConfig) configuration class: [FlexOlmoModel](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoModel) (FlexOlmoConfig model) - [FlintConfig](/docs/transformers/main/id/model_doc/flint#transformers.FlintConfig) configuration class: [FlintModel](/docs/transformers/main/id/model_doc/flint#transformers.FlintModel) (FlintConfig model) - [Florence2Config](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Config) configuration class: [Florence2Model](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Model) (Florence2Config model) - [FocalNetConfig](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetConfig) configuration class: [FocalNetModel](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetModel) (FocalNetConfig model) - [FunAsrNanoConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoConfig) configuration class: [FunAsrNanoModel](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoModel) (FunAsrNanoConfig model) - [FunAsrNanoEncoderConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoEncoderConfig) configuration class: [FunAsrNanoEncoder](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoEncoder) (FunAsrNanoEncoderConfig model) - [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig) configuration class: [FunnelModel](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelModel) or [FunnelBaseModel](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelBaseModel) (FunnelConfig model) - [FuyuConfig](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuConfig) configuration class: [FuyuModel](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuModel) (FuyuConfig model) - [GLPNConfig](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNConfig) configuration class: [GLPNModel](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNModel) (GLPNConfig model) - [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) configuration class: [GPT2Model](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Model) (GPT2Config model) - [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) configuration class: [GPTBigCodeModel](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeModel) (GPTBigCodeConfig model) - [GPTJConfig](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJConfig) configuration class: [GPTJModel](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJModel) (GPTJConfig model) - [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig) configuration class: [GPTNeoModel](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoModel) (GPTNeoConfig model) - [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig) configuration class: [GPTNeoXModel](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXModel) (GPTNeoXConfig model) - [GPTNeoXJapaneseConfig](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseConfig) configuration class: [GPTNeoXJapaneseModel](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseModel) (GPTNeoXJapaneseConfig model) - [Gemma2Config](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Config) configuration class: [Gemma2Model](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Model) (Gemma2Config model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3Model](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Model) (Gemma3Config model) - [Gemma3TextConfig](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextConfig) configuration class: [Gemma3TextModel](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextModel) (Gemma3TextConfig model) - [Gemma3nAudioConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nAudioConfig) configuration class: `Gemma3nAudioEncoder` (Gemma3nAudioConfig model) - [Gemma3nConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nConfig) configuration class: [Gemma3nModel](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nModel) (Gemma3nConfig model) - [Gemma3nTextConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nTextConfig) configuration class: [Gemma3nTextModel](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nTextModel) (Gemma3nTextConfig model) - [Gemma3nVisionConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nVisionConfig) configuration class: [TimmWrapperModel](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperModel) (Gemma3nVisionConfig model) - [Gemma4AudioConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioConfig) configuration class: [Gemma4AudioModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioModel) (Gemma4AudioConfig model) - [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) configuration class: [Gemma4Model](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Model) (Gemma4Config model) - [Gemma4TextConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4TextConfig) configuration class: [Gemma4TextModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4TextModel) (Gemma4TextConfig model) - [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) configuration class: [Gemma4UnifiedModel](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedModel) (Gemma4UnifiedConfig model) - [Gemma4UnifiedTextConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedTextConfig) configuration class: [Gemma4UnifiedTextModel](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedTextModel) (Gemma4UnifiedTextConfig model) - [Gemma4VisionConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4VisionConfig) configuration class: [Gemma4VisionModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4VisionModel) (Gemma4VisionConfig model) - [GemmaConfig](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaConfig) configuration class: [GemmaModel](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaModel) (GemmaConfig model) - [GitConfig](/docs/transformers/main/id/model_doc/git#transformers.GitConfig) configuration class: [GitModel](/docs/transformers/main/id/model_doc/git#transformers.GitModel) (GitConfig model) - [Glm46VConfig](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VConfig) configuration class: [Glm46VModel](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VModel) (Glm46VConfig model) - [Glm4Config](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Config) configuration class: [Glm4Model](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Model) (Glm4Config model) - [Glm4MoeConfig](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeConfig) configuration class: [Glm4MoeModel](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeModel) (Glm4MoeConfig model) - [Glm4MoeLiteConfig](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteConfig) configuration class: [Glm4MoeLiteModel](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteModel) (Glm4MoeLiteConfig model) - [Glm4vConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vConfig) configuration class: [Glm4vModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vModel) (Glm4vConfig model) - [Glm4vMoeConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeConfig) configuration class: [Glm4vMoeModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeModel) (Glm4vMoeConfig model) - [Glm4vMoeTextConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeTextConfig) configuration class: [Glm4vMoeTextModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeTextModel) (Glm4vMoeTextConfig model) - [Glm4vMoeVisionConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeVisionConfig) configuration class: [Glm4vMoeVisionModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeVisionModel) (Glm4vMoeVisionConfig model) - [Glm4vTextConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vTextConfig) configuration class: [Glm4vTextModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vTextModel) (Glm4vTextConfig model) - [Glm4vVisionConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vVisionConfig) configuration class: [Glm4vVisionModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vVisionModel) (Glm4vVisionConfig model) - [Glm5NextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextConfig) configuration class: [Glm5NextModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextModel) (Glm5NextConfig model) - [Glm5NextTextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextTextConfig) configuration class: [Glm5NextTextModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextTextModel) (Glm5NextTextConfig model) - [Glm5NextVisionConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextVisionConfig) configuration class: [Glm5NextVisionModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextVisionModel) (Glm5NextVisionConfig model) - [GlmAsrConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrConfig) configuration class: [GlmAsrModel](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrModel) (GlmAsrConfig model) - [GlmAsrEncoderConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrEncoderConfig) configuration class: [GlmAsrEncoder](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrEncoder) (GlmAsrEncoderConfig model) - [GlmConfig](/docs/transformers/main/id/model_doc/glm#transformers.GlmConfig) configuration class: [GlmModel](/docs/transformers/main/id/model_doc/glm#transformers.GlmModel) (GlmConfig model) - [GlmImageConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageConfig) configuration class: [GlmImageModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageModel) (GlmImageConfig model) - [GlmImageTextConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageTextConfig) configuration class: [GlmImageTextModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageTextModel) (GlmImageTextConfig model) - [GlmImageVQVAEConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVQVAEConfig) configuration class: [GlmImageVQVAE](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVQVAE) (GlmImageVQVAEConfig model) - [GlmImageVisionConfig](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVisionConfig) configuration class: [GlmImageVisionModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVisionModel) (GlmImageVisionConfig model) - [GlmMoeDsaConfig](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaConfig) configuration class: [GlmMoeDsaModel](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaModel) (GlmMoeDsaConfig model) - [GlmOcrConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrConfig) configuration class: [GlmOcrModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrModel) (GlmOcrConfig model) - [GlmOcrTextConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrTextConfig) configuration class: [GlmOcrTextModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrTextModel) (GlmOcrTextConfig model) - [GlmOcrVisionConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrVisionConfig) configuration class: [GlmOcrVisionModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrVisionModel) (GlmOcrVisionConfig model) - [GlmgaConfig](/docs/transformers/main/id/model_doc/glmga#transformers.GlmgaConfig) configuration class: [Glm46VModel](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VModel) (GlmgaConfig model) - [GotOcr2Config](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Config) configuration class: [GotOcr2Model](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Model) (GotOcr2Config model) - [GptOssConfig](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssConfig) configuration class: [GptOssModel](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssModel) (GptOssConfig model) - [GradiumPhononTTSConfig](/docs/transformers/main/id/model_doc/gradium_phonon_tts#transformers.GradiumPhononTTSConfig) configuration class: [GradiumPhononTTSModel](/docs/transformers/main/id/model_doc/gradium_phonon_tts#transformers.GradiumPhononTTSModel) (GradiumPhononTTSConfig model) - [Granite4VisionConfig](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionConfig) configuration class: [Granite4VisionModel](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionModel) (Granite4VisionConfig model) - [GraniteConfig](/docs/transformers/main/id/model_doc/granite#transformers.GraniteConfig) configuration class: [GraniteModel](/docs/transformers/main/id/model_doc/granite#transformers.GraniteModel) (GraniteConfig model) - [GraniteMoeConfig](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeConfig) configuration class: [GraniteMoeModel](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeModel) (GraniteMoeConfig model) - [GraniteMoeHybridConfig](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridConfig) configuration class: [GraniteMoeHybridModel](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridModel) (GraniteMoeHybridConfig model) - [GraniteMoeSWAConfig](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAConfig) configuration class: [GraniteMoeSWAModel](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAModel) (GraniteMoeSWAConfig model) - [GraniteMoeSharedConfig](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedConfig) configuration class: [GraniteMoeSharedModel](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedModel) (GraniteMoeSharedConfig model) - [GraniteSWAConfig](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAConfig) configuration class: [GraniteSWAModel](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAModel) (GraniteSWAConfig model) - [GraniteSpeech5CTCConfig](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5CTCConfig) configuration class: [GraniteSpeech5ForCTC](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5ForCTC) (GraniteSpeech5CTCConfig model) - [GraniteSpeech5EncoderConfig](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5EncoderConfig) configuration class: [GraniteSpeech5Encoder](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5Encoder) (GraniteSpeech5EncoderConfig model) - [GraniteSpeechConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechConfig) configuration class: [GraniteSpeechModel](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechModel) (GraniteSpeechConfig model) - [GraniteSpeechPlusConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusConfig) configuration class: [GraniteSpeechPlusModel](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusModel) (GraniteSpeechPlusConfig model) - [GroundingDinoConfig](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoConfig) configuration class: [GroundingDinoModel](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoModel) (GroundingDinoConfig model) - [GroupViTConfig](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTConfig) configuration class: [GroupViTModel](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTModel) (GroupViTConfig model) - [GteConfig](/docs/transformers/main/id/model_doc/gte#transformers.GteConfig) configuration class: [GteModel](/docs/transformers/main/id/model_doc/gte#transformers.GteModel) (GteConfig model) - [HGNetV2Config](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2Config) configuration class: [HGNetV2Backbone](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2Backbone) (HGNetV2Config model) - [HYV3Config](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3Config) configuration class: [HYV3Model](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3Model) (HYV3Config model) - [HYV4Config](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4Config) configuration class: [HYV4Model](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4Model) (HYV4Config model) - [HeliumConfig](/docs/transformers/main/id/model_doc/helium#transformers.HeliumConfig) configuration class: [HeliumModel](/docs/transformers/main/id/model_doc/helium#transformers.HeliumModel) (HeliumConfig model) - [HieraConfig](/docs/transformers/main/id/model_doc/hiera#transformers.HieraConfig) configuration class: [HieraModel](/docs/transformers/main/id/model_doc/hiera#transformers.HieraModel) (HieraConfig model) - [HiggsAudioV2Config](/docs/transformers/main/id/model_doc/higgs_audio_v2#transformers.HiggsAudioV2Config) configuration class: [HiggsAudioV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/higgs_audio_v2#transformers.HiggsAudioV2ForConditionalGeneration) (HiggsAudioV2Config model) - [HiggsAudioV2TokenizerConfig](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerConfig) configuration class: [HiggsAudioV2TokenizerModel](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerModel) (HiggsAudioV2TokenizerConfig model) - [HrmTextConfig](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextConfig) configuration class: [HrmTextModel](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextModel) (HrmTextConfig model) - [HubertConfig](/docs/transformers/main/id/model_doc/hubert#transformers.HubertConfig) configuration class: [HubertModel](/docs/transformers/main/id/model_doc/hubert#transformers.HubertModel) (HubertConfig model) - [HunYuanDenseV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Config) configuration class: [HunYuanDenseV1Model](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Model) (HunYuanDenseV1Config model) - [HunYuanMoEV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Config) configuration class: [HunYuanMoEV1Model](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Model) (HunYuanMoEV1Config model) - [HunYuanVLConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLConfig) configuration class: [HunYuanVLModel](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLModel) (HunYuanVLConfig model) - [HunYuanVLTextConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLTextConfig) configuration class: [HunYuanVLTextModel](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLTextModel) (HunYuanVLTextConfig model) - [HunYuanVLVisionConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLVisionConfig) configuration class: [HunYuanVLVisionTransformer](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLVisionTransformer) (HunYuanVLVisionConfig model) - [HyperCLOVAXConfig](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXConfig) configuration class: [HyperCLOVAXModel](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXModel) (HyperCLOVAXConfig model) - [HyperCLOVAXVisionV2Config](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Config) configuration class: [HyperCLOVAXVisionV2Model](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Model) (HyperCLOVAXVisionV2Config model) - [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig) configuration class: [IBertModel](/docs/transformers/main/id/model_doc/ibert#transformers.IBertModel) (IBertConfig model) - [IJepaConfig](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaConfig) configuration class: [IJepaModel](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaModel) (IJepaConfig model) - [Idefics2Config](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Config) configuration class: [Idefics2Model](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Model) (Idefics2Config model) - [Idefics3Config](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Config) configuration class: [Idefics3Model](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Model) (Idefics3Config model) - [Idefics3VisionConfig](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3VisionConfig) configuration class: [Idefics3VisionTransformer](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3VisionTransformer) (Idefics3VisionConfig model) - [IdeficsConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsConfig) configuration class: [IdeficsModel](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsModel) (IdeficsConfig model) - [ImageGPTConfig](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTConfig) configuration class: [ImageGPTModel](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTModel) (ImageGPTConfig model) - [InformerConfig](/docs/transformers/main/id/model_doc/informer#transformers.InformerConfig) configuration class: [InformerModel](/docs/transformers/main/id/model_doc/informer#transformers.InformerModel) (InformerConfig model) - [InklingConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingConfig) configuration class: [InklingModel](/docs/transformers/main/id/model_doc/inkling#transformers.InklingModel) (InklingConfig model) - [InklingTextConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingTextConfig) configuration class: [InklingTextModel](/docs/transformers/main/id/model_doc/inkling#transformers.InklingTextModel) (InklingTextConfig model) - [InstructBlipConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipConfig) configuration class: [InstructBlipModel](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipModel) (InstructBlipConfig model) - [InstructBlipVideoConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoConfig) configuration class: [InstructBlipVideoModel](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoModel) (InstructBlipVideoConfig model) - [InternVLConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLConfig) configuration class: [InternVLModel](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLModel) (InternVLConfig model) - [InternVLVisionConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLVisionConfig) configuration class: [InternVLVisionModel](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLVisionModel) (InternVLVisionConfig model) - [Jais2Config](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2Config) configuration class: [Jais2Model](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2Model) (Jais2Config model) - [JambaConfig](/docs/transformers/main/id/model_doc/jamba#transformers.JambaConfig) configuration class: [JambaModel](/docs/transformers/main/id/model_doc/jamba#transformers.JambaModel) (JambaConfig model) - [JanusConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusConfig) configuration class: [JanusModel](/docs/transformers/main/id/model_doc/janus#transformers.JanusModel) (JanusConfig model) - [JetMoeConfig](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeConfig) configuration class: [JetMoeModel](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeModel) (JetMoeConfig model) - [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config) configuration class: [JinaEmbeddingsV3Model](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Model) (JinaEmbeddingsV3Config model) - [KimiLinearConfig](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearConfig) configuration class: [KimiLinearModel](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearModel) (KimiLinearConfig model) - [Kimi_K25Config](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Config) configuration class: [Kimi_K25Model](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Model) (Kimi_K25Config model) - [Kimi_K25VisionConfig](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25VisionConfig) configuration class: [Kimi_K25VisionModel](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25VisionModel) (Kimi_K25VisionConfig model) - [Kolibri1Config](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1Config) configuration class: [Kolibri1Model](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1Model) (Kolibri1Config model) - [Kosmos2Config](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Config) configuration class: [Kosmos2Model](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Model) (Kosmos2Config model) - [Kosmos2_5Config](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Config) configuration class: [Kosmos2_5Model](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Model) (Kosmos2_5Config model) - [KyutaiSpeechToTextConfig](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextConfig) configuration class: [KyutaiSpeechToTextModel](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextModel) (KyutaiSpeechToTextConfig model) - [LEDConfig](/docs/transformers/main/id/model_doc/led#transformers.LEDConfig) configuration class: [LEDModel](/docs/transformers/main/id/model_doc/led#transformers.LEDModel) (LEDConfig model) - [LagunaConfig](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaConfig) configuration class: [LagunaModel](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaModel) (LagunaConfig model) - [LasrCTCConfig](/docs/transformers/main/id/model_doc/lasr#transformers.LasrCTCConfig) configuration class: [LasrForCTC](/docs/transformers/main/id/model_doc/lasr#transformers.LasrForCTC) (LasrCTCConfig model) - [LasrEncoderConfig](/docs/transformers/main/id/model_doc/lasr#transformers.LasrEncoderConfig) configuration class: [LasrEncoder](/docs/transformers/main/id/model_doc/lasr#transformers.LasrEncoder) (LasrEncoderConfig model) - [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig) configuration class: [LayoutLMModel](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMModel) (LayoutLMConfig model) - [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config) configuration class: [LayoutLMv2Model](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Model) (LayoutLMv2Config model) - [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config) configuration class: [LayoutLMv3Model](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Model) (LayoutLMv3Config model) - [LevitConfig](/docs/transformers/main/id/model_doc/levit#transformers.LevitConfig) configuration class: [LevitModel](/docs/transformers/main/id/model_doc/levit#transformers.LevitModel) (LevitConfig model) - [Lfm2Config](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2Config) configuration class: [Lfm2Model](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2Model) (Lfm2Config model) - [Lfm2MoeConfig](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeConfig) configuration class: [Lfm2MoeModel](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeModel) (Lfm2MoeConfig model) - [Lfm2VlConfig](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlConfig) configuration class: [Lfm2VlModel](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlModel) (Lfm2VlConfig model) - [Lfm3Config](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3Config) configuration class: [Lfm3Model](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3Model) (Lfm3Config model) - [LightGlueConfig](/docs/transformers/main/id/model_doc/lightglue#transformers.LightGlueConfig) configuration class: [LightGlueForKeypointMatching](/docs/transformers/main/id/model_doc/lightglue#transformers.LightGlueForKeypointMatching) (LightGlueConfig model) - [LightOnOcrConfig](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrConfig) configuration class: [LightOnOcrModel](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrModel) (LightOnOcrConfig model) - [LiltConfig](/docs/transformers/main/id/model_doc/lilt#transformers.LiltConfig) configuration class: [LiltModel](/docs/transformers/main/id/model_doc/lilt#transformers.LiltModel) (LiltConfig model) - [Llama4Config](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Config) configuration class: [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (Llama4Config model) - [Llama4TextConfig](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4TextConfig) configuration class: [Llama4TextModel](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4TextModel) (Llama4TextConfig model) - [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig) configuration class: [LlamaModel](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaModel) (LlamaConfig model) - [LlavaConfig](/docs/transformers/main/id/model_doc/llava#transformers.LlavaConfig) configuration class: [LlavaModel](/docs/transformers/main/id/model_doc/llava#transformers.LlavaModel) (LlavaConfig model) - [LlavaNextConfig](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextConfig) configuration class: [LlavaNextModel](/docs/transformers/main/id/model_doc/llava_next#transformers.LlavaNextModel) (LlavaNextConfig model) - [LlavaNextVideoConfig](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoConfig) configuration class: [LlavaNextVideoModel](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoModel) (LlavaNextVideoConfig model) - [LlavaOnevisionConfig](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionConfig) configuration class: [LlavaOnevisionModel](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionModel) (LlavaOnevisionConfig model) - [LongT5Config](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5Config) configuration class: [LongT5Model](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5Model) (LongT5Config model) - [LongcatFlashConfig](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashConfig) configuration class: [LongcatFlashModel](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashModel) (LongcatFlashConfig model) - [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig) configuration class: [LongformerModel](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerModel) (LongformerConfig model) - [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig) configuration class: [LukeModel](/docs/transformers/main/id/model_doc/luke#transformers.LukeModel) (LukeConfig model) - [LwDetrConfig](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrConfig) configuration class: [LwDetrModel](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrModel) (LwDetrConfig model) - [LxmertConfig](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertConfig) configuration class: [LxmertModel](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertModel) (LxmertConfig model) - [M2M100Config](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Config) configuration class: [M2M100Model](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Model) (M2M100Config model) - [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig) configuration class: [MBartModel](/docs/transformers/main/id/model_doc/mbart#transformers.MBartModel) (MBartConfig model) - [MLCDVisionConfig](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionConfig) configuration class: [MLCDVisionModel](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionModel) (MLCDVisionConfig model) - [MMGroundingDinoConfig](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoConfig) configuration class: [MMGroundingDinoModel](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoModel) (MMGroundingDinoConfig model) - [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig) configuration class: [MPNetModel](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetModel) (MPNetConfig model) - [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config) configuration class: [MT5Model](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Model) (MT5Config model) - [Mamba2Config](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Config) configuration class: [Mamba2Model](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Model) (Mamba2Config model) - [MambaConfig](/docs/transformers/main/id/model_doc/mamba#transformers.MambaConfig) configuration class: [MambaModel](/docs/transformers/main/id/model_doc/mamba#transformers.MambaModel) (MambaConfig model) - [MarianConfig](/docs/transformers/main/id/model_doc/marian#transformers.MarianConfig) configuration class: [MarianModel](/docs/transformers/main/id/model_doc/marian#transformers.MarianModel) (MarianConfig model) - [MarkupLMConfig](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMConfig) configuration class: [MarkupLMModel](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMModel) (MarkupLMConfig model) - [Mask2FormerConfig](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerConfig) configuration class: [Mask2FormerModel](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerModel) (Mask2FormerConfig model) - [MaskFormerConfig](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerConfig) configuration class: [MaskFormerModel](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerModel) (MaskFormerConfig model) - `MaskFormerSwinConfig` configuration class: `MaskFormerSwinModel` (MaskFormerSwinConfig model) - [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) configuration class: [MegatronBertModel](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertModel) (MegatronBertConfig model) - [MellumConfig](/docs/transformers/main/id/model_doc/mellum#transformers.MellumConfig) configuration class: [MellumModel](/docs/transformers/main/id/model_doc/mellum#transformers.MellumModel) (MellumConfig model) - [MetaClip2Config](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Config) configuration class: [MetaClip2Model](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Model) (MetaClip2Config model) - [MgpstrConfig](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrConfig) configuration class: [MgpstrForSceneTextRecognition](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrForSceneTextRecognition) (MgpstrConfig model) - [MiMoV2FlashConfig](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashConfig) configuration class: [MiMoV2FlashModel](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashModel) (MiMoV2FlashConfig model) - [MimiConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiConfig) configuration class: [MimiModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiModel) (MimiConfig model) - [MimiDecoderConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiDecoderConfig) configuration class: [MimiDecoderModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiDecoderModel) (MimiDecoderConfig model) - [MimiEncoderConfig](/docs/transformers/main/id/model_doc/mimi#transformers.MimiEncoderConfig) configuration class: [MimiEncoderModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiEncoderModel) (MimiEncoderConfig model) - [MiniCPM3Config](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Config) configuration class: [MiniCPM3Model](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Model) (MiniCPM3Config model) - [MiniCPMOcrConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrConfig) configuration class: [MiniCPMOcrModel](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrModel) (MiniCPMOcrConfig model) - [MiniCPMOcrVisionConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrVisionConfig) configuration class: [MiniCPMOcrVisionModel](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrVisionModel) (MiniCPMOcrVisionConfig model) - [MiniCPMV4_6Config](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Config) configuration class: [MiniCPMV4_6Model](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Model) (MiniCPMV4_6Config model) - [MiniCPMV4_7Config](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Config) configuration class: [MiniCPMV4_7Model](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Model) (MiniCPMV4_7Config model) - [MiniCPMV4_7VisionConfig](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7VisionConfig) configuration class: [MiniCPMV4_7VisionModel](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7VisionModel) (MiniCPMV4_7VisionConfig model) - [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig) configuration class: [MiniMaxModel](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxModel) (MiniMaxConfig model) - [MiniMaxM2Config](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2Config) configuration class: [MiniMaxM2Model](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2Model) (MiniMaxM2Config model) - [MiniMaxM3VLConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLConfig) configuration class: [MiniMaxM3VLModel](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLModel) (MiniMaxM3VLConfig model) - [MiniMaxM3VLTextConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLTextConfig) configuration class: [MiniMaxM3VLTextModel](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLTextModel) (MiniMaxM3VLTextConfig model) - [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config) configuration class: [Ministral3Model](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Model) (Ministral3Config model) - [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig) configuration class: [MinistralModel](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralModel) (MinistralConfig model) - [Mistral3Config](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Config) configuration class: [Mistral3Model](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Model) (Mistral3Config model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4Model](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Model) (Mistral4Config model) - [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig) configuration class: [MistralModel](/docs/transformers/main/id/model_doc/mistral#transformers.MistralModel) (MistralConfig model) - [MistralLarge4Config](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Config) configuration class: [MistralLarge4Model](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Model) (MistralLarge4Config model) - [MistralLarge4TextConfig](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4TextConfig) configuration class: [MistralLarge4TextModel](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4TextModel) (MistralLarge4TextConfig model) - [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig) configuration class: [MixtralModel](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralModel) (MixtralConfig model) - [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) configuration class: [MllamaModel](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaModel) (MllamaConfig model) - [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig) configuration class: [MobileBertModel](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertModel) (MobileBertConfig model) - [MobileNetV1Config](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1Config) configuration class: [MobileNetV1Model](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1Model) (MobileNetV1Config model) - [MobileNetV2Config](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Config) configuration class: [MobileNetV2Model](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Model) (MobileNetV2Config model) - [MobileViTConfig](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTConfig) configuration class: [MobileViTModel](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTModel) (MobileViTConfig model) - [MobileViTV2Config](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Config) configuration class: [MobileViTV2Model](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Model) (MobileViTV2Config model) - [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig) configuration class: [ModernBertModel](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertModel) (ModernBertConfig model) - [ModernBertDecoderConfig](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderConfig) configuration class: [ModernBertDecoderModel](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderModel) (ModernBertDecoderConfig model) - [ModernVBertConfig](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertConfig) configuration class: [ModernVBertModel](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertModel) (ModernVBertConfig model) - [Molmo2Config](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Config) configuration class: [Molmo2Model](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Model) (Molmo2Config model) - [Molmo2TextConfig](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2TextConfig) configuration class: [Molmo2TextModel](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2TextModel) (Molmo2TextConfig model) - [MoonshineConfig](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineConfig) configuration class: [MoonshineModel](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineModel) (MoonshineConfig model) - [MoonshineStreamingConfig](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingConfig) configuration class: [MoonshineStreamingModel](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingModel) (MoonshineStreamingConfig model) - [MoshiConfig](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiConfig) configuration class: [MoshiModel](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiModel) (MoshiConfig model) - [MossTranscribeDiarizeAudioConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeAudioConfig) configuration class: [MossTranscribeDiarizeAudioModel](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeAudioModel) (MossTranscribeDiarizeAudioConfig model) - [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig) configuration class: [MossTranscribeDiarizeModel](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeModel) (MossTranscribeDiarizeConfig model) - [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) configuration class: [MptModel](/docs/transformers/main/id/model_doc/mpt#transformers.MptModel) (MptConfig model) - [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig) configuration class: [MraModel](/docs/transformers/main/id/model_doc/mra#transformers.MraModel) (MraConfig model) - [MuseGlimmerAssistantConfig](/docs/transformers/main/id/model_doc/muse_glimmer_assistant#transformers.MuseGlimmerAssistantConfig) configuration class: [MuseGlimmerAssistantModel](/docs/transformers/main/id/model_doc/muse_glimmer_assistant#transformers.MuseGlimmerAssistantModel) (MuseGlimmerAssistantConfig model) - [MuseGlimmerConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerConfig) configuration class: [MuseGlimmerModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerModel) (MuseGlimmerConfig model) - [MuseGlimmerTextConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerTextConfig) configuration class: [MuseGlimmerTextModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerTextModel) (MuseGlimmerTextConfig model) - [MuseGlimmerVisionConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerVisionConfig) configuration class: [MuseGlimmerVisionModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerVisionModel) (MuseGlimmerVisionConfig model) - [MuseSparkAudioConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkAudioConfig) configuration class: [MuseSparkAudioModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkAudioModel) (MuseSparkAudioConfig model) - [MuseSparkConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkConfig) configuration class: [MuseSparkModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkModel) (MuseSparkConfig model) - [MuseSparkTextConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkTextConfig) configuration class: [MuseSparkTextModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkTextModel) (MuseSparkTextConfig model) - [MuseSparkVisionConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkVisionConfig) configuration class: [MuseSparkVisionModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkVisionModel) (MuseSparkVisionConfig model) - [MusicFlamingoConfig](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoConfig) configuration class: [MusicFlamingoModel](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoModel) (MusicFlamingoConfig model) - [MusicgenConfig](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenConfig) configuration class: [MusicgenModel](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenModel) (MusicgenConfig model) - [MusicgenMelodyConfig](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyConfig) configuration class: [MusicgenMelodyModel](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyModel) (MusicgenMelodyConfig model) - [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig) configuration class: [MvpModel](/docs/transformers/main/id/model_doc/mvp#transformers.MvpModel) (MvpConfig model) - [NanoChatConfig](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatConfig) configuration class: [NanoChatModel](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatModel) (NanoChatConfig model) - [Nemotron3DiarizationAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationAudioConfig) configuration class: [Nemotron3DiarizationAudioModel](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationAudioModel) (Nemotron3DiarizationAudioConfig model) - [Nemotron3DiarizationConfig](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationConfig) configuration class: [Nemotron3DiarizationModel](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationModel) (Nemotron3DiarizationConfig model) - [Nemotron3_5AsrConfig](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrConfig) configuration class: [Nemotron3_5AsrForRNNT](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrForRNNT) (Nemotron3_5AsrConfig model) - [Nemotron3_5TranscribeAsrAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAsrAudioConfig) configuration class: [Nemotron3_5TranscribeAsrAudioModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAsrAudioModel) (Nemotron3_5TranscribeAsrAudioConfig model) - [Nemotron3_5TranscribeAudioConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAudioConfig) configuration class: [Nemotron3_5TranscribeAudioModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAudioModel) (Nemotron3_5TranscribeAudioConfig model) - [Nemotron3_5TranscribeConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeConfig) configuration class: [Nemotron3_5TranscribeModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeModel) (Nemotron3_5TranscribeConfig model) - [Nemotron3_5TranscribeWordTimestampConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeWordTimestampConfig) configuration class: [Nemotron3_5TranscribeWordTimestampModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeWordTimestampModel) (Nemotron3_5TranscribeWordTimestampConfig model) - [NemotronAsrStreamingConfig](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingConfig) configuration class: [NemotronAsrStreamingForRNNT](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingForRNNT) (NemotronAsrStreamingConfig model) - [NemotronAsrStreamingEncoderConfig](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingEncoderConfig) configuration class: [NemotronAsrStreamingEncoder](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingEncoder) (NemotronAsrStreamingEncoderConfig model) - [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig) configuration class: [NemotronModel](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronModel) (NemotronConfig model) - [NemotronHConfig](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHConfig) configuration class: [NemotronHModel](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHModel) (NemotronHConfig model) - [NeoMMEConfig](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEConfig) configuration class: [NeoMMEModel](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEModel) (NeoMMEConfig model) - [NeuCodecConfig](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecConfig) configuration class: [NeuCodecModel](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecModel) (NeuCodecConfig model) - [NllbMoeConfig](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeConfig) configuration class: [NllbMoeModel](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeModel) (NllbMoeConfig model) - [NomicBertConfig](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertConfig) configuration class: [NomicBertModel](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertModel) (NomicBertConfig model) - [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig) configuration class: [NystromformerModel](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerModel) (NystromformerConfig model) - [OPTConfig](/docs/transformers/main/id/model_doc/opt#transformers.OPTConfig) configuration class: [OPTModel](/docs/transformers/main/id/model_doc/opt#transformers.OPTModel) (OPTConfig model) - [Olmo2Config](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Config) configuration class: [Olmo2Model](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Model) (Olmo2Config model) - [Olmo3Config](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Config) configuration class: [Olmo3Model](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Model) (Olmo3Config model) - [OlmoConfig](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoConfig) configuration class: [OlmoModel](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoModel) (OlmoConfig model) - [OlmoHybridConfig](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridConfig) configuration class: [OlmoHybridModel](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridModel) (OlmoHybridConfig model) - [OlmoeConfig](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeConfig) configuration class: [OlmoeModel](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeModel) (OlmoeConfig model) - [OmDetTurboConfig](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboConfig) configuration class: [OmDetTurboForObjectDetection](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboForObjectDetection) (OmDetTurboConfig model) - [OneFormerConfig](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerConfig) configuration class: [OneFormerModel](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerModel) (OneFormerConfig model) - [OpenAIGPTConfig](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTConfig) configuration class: [OpenAIGPTModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTModel) (OpenAIGPTConfig model) - [OpenAIPrivacyFilterConfig](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterConfig) configuration class: [OpenAIPrivacyFilterModel](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterModel) (OpenAIPrivacyFilterConfig model) - [Ovis2Config](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Config) configuration class: [Ovis2Model](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Model) (Ovis2Config model) - [OwlViTConfig](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTConfig) configuration class: [OwlViTModel](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTModel) (OwlViTConfig model) - [Owlv2Config](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Config) configuration class: [Owlv2Model](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Model) (Owlv2Config model) - [PI0Config](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Config) configuration class: [PI0Model](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Model) (PI0Config model) - [PLBartConfig](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartConfig) configuration class: [PLBartModel](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartModel) (PLBartConfig model) - [PPDocLayoutV3Config](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3Config) configuration class: [PPDocLayoutV3Model](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3Model) (PPDocLayoutV3Config model) - [PPDocLayoutV4Config](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4Config) configuration class: [PPDocLayoutV4Model](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4Model) (PPDocLayoutV4Config model) - [PPOCRV5MobileRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecConfig) configuration class: [PPOCRV5MobileRecModel](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecModel) (PPOCRV5MobileRecConfig model) - [PPOCRV5ServerRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecConfig) configuration class: [PPOCRV5ServerRecModel](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecModel) (PPOCRV5ServerRecConfig model) - [PPOCRV6MediumDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetConfig) configuration class: [PPOCRV6MediumDetModel](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetModel) (PPOCRV6MediumDetConfig model) - [PPOCRV6SmallDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetConfig) configuration class: [PPOCRV6SmallDetModel](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetModel) (PPOCRV6SmallDetConfig model) - [PPOCRV6SmallRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecConfig) configuration class: [PPOCRV6SmallRecModel](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecModel) (PPOCRV6SmallRecConfig model) - [PPOCRV6TinyRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecConfig) configuration class: [PPOCRV6TinyRecModel](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecModel) (PPOCRV6TinyRecConfig model) - [PaddleOCRVLConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLConfig) configuration class: [PaddleOCRVLModel](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLModel) (PaddleOCRVLConfig model) - [PaliGemmaConfig](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaConfig) configuration class: [PaliGemmaModel](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaModel) (PaliGemmaConfig model) - [ParakeetCTCConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetCTCConfig) configuration class: [ParakeetForCTC](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForCTC) (ParakeetCTCConfig model) - [ParakeetEncoderConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetEncoderConfig) configuration class: [ParakeetEncoder](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetEncoder) (ParakeetEncoderConfig model) - [ParakeetRNNTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetRNNTConfig) configuration class: [ParakeetForRNNT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForRNNT) (ParakeetRNNTConfig model) - [ParakeetTDTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTDTConfig) configuration class: [ParakeetForTDT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForTDT) (ParakeetTDTConfig model) - [PatchTSMixerConfig](/docs/transformers/main/id/model_doc/patchtsmixer#transformers.PatchTSMixerConfig) configuration class: [PatchTSMixerModel](/docs/transformers/main/id/model_doc/patchtsmixer#transformers.PatchTSMixerModel) (PatchTSMixerConfig model) - [PatchTSTConfig](/docs/transformers/main/id/model_doc/patchtst#transformers.PatchTSTConfig) configuration class: [PatchTSTModel](/docs/transformers/main/id/model_doc/patchtst#transformers.PatchTSTModel) (PatchTSTConfig model) - [PeAudioConfig](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioConfig) configuration class: [PeAudioModel](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioModel) (PeAudioConfig model) - [PeAudioEncoderConfig](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioEncoderConfig) configuration class: [PeAudioEncoder](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioEncoder) (PeAudioEncoderConfig model) - [PeAudioVideoConfig](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoConfig) configuration class: [PeAudioVideoModel](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoModel) (PeAudioVideoConfig model) - [PeAudioVideoEncoderConfig](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoEncoderConfig) configuration class: [PeAudioVideoEncoder](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoEncoder) (PeAudioVideoEncoderConfig model) - [PeVideoConfig](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoConfig) configuration class: [PeVideoModel](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoModel) (PeVideoConfig model) - [PeVideoEncoderConfig](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoEncoderConfig) configuration class: [PeVideoEncoder](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoEncoder) (PeVideoEncoderConfig model) - [PegasusConfig](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusConfig) configuration class: [PegasusModel](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusModel) (PegasusConfig model) - [PegasusXConfig](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXConfig) configuration class: [PegasusXModel](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXModel) (PegasusXConfig model) - [PerceiverConfig](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverConfig) configuration class: [PerceiverModel](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverModel) (PerceiverConfig model) - [PerceptionLMConfig](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMConfig) configuration class: [PerceptionLMModel](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMModel) (PerceptionLMConfig model) - [PersimmonConfig](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonConfig) configuration class: [PersimmonModel](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonModel) (PersimmonConfig model) - [Phi3Config](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Config) configuration class: [Phi3Model](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Model) (Phi3Config model) - [Phi4MultimodalConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalConfig) configuration class: [Phi4MultimodalModel](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalModel) (Phi4MultimodalConfig model) - [PhiConfig](/docs/transformers/main/id/model_doc/phi#transformers.PhiConfig) configuration class: [PhiModel](/docs/transformers/main/id/model_doc/phi#transformers.PhiModel) (PhiConfig model) - [PhimoeConfig](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeConfig) configuration class: [PhimoeModel](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeModel) (PhimoeConfig model) - [PixioConfig](/docs/transformers/main/id/model_doc/pixio#transformers.PixioConfig) configuration class: [PixioModel](/docs/transformers/main/id/model_doc/pixio#transformers.PixioModel) (PixioConfig model) - [PixtralVisionConfig](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralVisionConfig) configuration class: [PixtralVisionModel](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralVisionModel) (PixtralVisionConfig model) - [PocketTTSAudioTokenizerConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerConfig) configuration class: [PocketTTSAudioTokenizerModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerModel) (PocketTTSAudioTokenizerConfig model) - [PocketTTSAudioTokenizerDecoderConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerDecoderConfig) configuration class: [PocketTTSAudioTokenizerDecoderModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerDecoderModel) (PocketTTSAudioTokenizerDecoderConfig model) - [PocketTTSAudioTokenizerEncoderConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerEncoderConfig) configuration class: [PocketTTSAudioTokenizerEncoderModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerEncoderModel) (PocketTTSAudioTokenizerEncoderConfig model) - [PocketTTSConfig](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSConfig) configuration class: [PocketTTSModel](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSModel) (PocketTTSConfig model) - [PoolFormerConfig](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerConfig) configuration class: [PoolFormerModel](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerModel) (PoolFormerConfig model) - [ProphetNetConfig](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetConfig) configuration class: [ProphetNetModel](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetModel) (ProphetNetConfig model) - [PvtConfig](/docs/transformers/main/id/model_doc/pvt#transformers.PvtConfig) configuration class: [PvtModel](/docs/transformers/main/id/model_doc/pvt#transformers.PvtModel) (PvtConfig model) - [PvtV2Config](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2Config) configuration class: [PvtV2Model](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2Model) (PvtV2Config model) - [QianfanOCRConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRConfig) configuration class: [QianfanOCRModel](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRModel) (QianfanOCRConfig model) - [QianfanOCRVisionConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRVisionConfig) configuration class: [QianfanOCRVisionModel](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRVisionModel) (QianfanOCRVisionConfig model) - [Qwen2AudioConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioConfig) configuration class: [Qwen2AudioModel](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioModel) (Qwen2AudioConfig model) - [Qwen2AudioEncoderConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioEncoderConfig) configuration class: [Qwen2AudioEncoder](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioEncoder) (Qwen2AudioEncoderConfig model) - [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config) configuration class: [Qwen2Model](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Model) (Qwen2Config model) - [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig) configuration class: [Qwen2MoeModel](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeModel) (Qwen2MoeConfig model) - [Qwen2VLConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLConfig) configuration class: [Qwen2VLModel](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLModel) (Qwen2VLConfig model) - [Qwen2VLTextConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLTextConfig) configuration class: [Qwen2VLTextModel](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLTextModel) (Qwen2VLTextConfig model) - [Qwen2_5OmniThinkerConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerConfig) configuration class: [Qwen2_5OmniThinkerModel](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerModel) (Qwen2_5OmniThinkerConfig model) - [Qwen2_5_VLConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLConfig) configuration class: [Qwen2_5_VLModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLModel) (Qwen2_5_VLConfig model) - [Qwen2_5_VLTextConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLTextConfig) configuration class: [Qwen2_5_VLTextModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLTextModel) (Qwen2_5_VLTextConfig model) - [Qwen2_5_VLVisionConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLVisionConfig) configuration class: [Qwen2_5_VisionTransformerPretrainedModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VisionTransformerPretrainedModel) (Qwen2_5_VLVisionConfig model) - [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig) configuration class: [Qwen3ASRModel](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRModel) (Qwen3ASRConfig model) - [Qwen3ASREncoderConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASREncoderConfig) configuration class: [Qwen3ASREncoder](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASREncoder) (Qwen3ASREncoderConfig model) - [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config) configuration class: [Qwen3Model](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Model) (Qwen3Config model) - [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig) configuration class: [Qwen3MoeModel](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeModel) (Qwen3MoeConfig model) - [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig) configuration class: [Qwen3NextModel](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextModel) (Qwen3NextConfig model) - [Qwen3OmniMoeThinkerConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerConfig) configuration class: [Qwen3OmniMoeThinkerModel](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerModel) (Qwen3OmniMoeThinkerConfig model) - [Qwen3VLConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLConfig) configuration class: [Qwen3VLModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLModel) (Qwen3VLConfig model) - [Qwen3VLMoeConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeConfig) configuration class: [Qwen3VLMoeModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeModel) (Qwen3VLMoeConfig model) - [Qwen3VLMoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeTextConfig) configuration class: [Qwen3VLMoeTextModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeTextModel) (Qwen3VLMoeTextConfig model) - [Qwen3VLMoeVisionConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeVisionConfig) configuration class: [Qwen3VLMoeVisionModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeVisionModel) (Qwen3VLMoeVisionConfig model) - [Qwen3VLTextConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLTextConfig) configuration class: [Qwen3VLTextModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLTextModel) (Qwen3VLTextConfig model) - [Qwen3VLVisionConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLVisionConfig) configuration class: [Qwen3VLVisionModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLVisionModel) (Qwen3VLVisionConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5Model](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Model) (Qwen3_5Config model) - [Qwen3_5MoeConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeConfig) configuration class: [Qwen3_5MoeModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeModel) (Qwen3_5MoeConfig model) - [Qwen3_5MoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeTextConfig) configuration class: [Qwen3_5MoeTextModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeTextModel) (Qwen3_5MoeTextConfig model) - [Qwen3_5MoeVisionConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeVisionConfig) configuration class: [Qwen3_5MoeVisionModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeVisionModel) (Qwen3_5MoeVisionConfig model) - [Qwen3_5TextConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextConfig) configuration class: [Qwen3_5TextModel](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextModel) (Qwen3_5TextConfig model) - [Qwen3_5VisionConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5VisionConfig) configuration class: [Qwen3_5VisionModel](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5VisionModel) (Qwen3_5VisionConfig model) - [Qwen4ExpConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpConfig) configuration class: [Qwen4ExpModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpModel) (Qwen4ExpConfig model) - [Qwen4ExpTextConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpTextConfig) configuration class: [Qwen4ExpTextModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpTextModel) (Qwen4ExpTextConfig model) - [Qwen4ExpVisionConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpVisionConfig) configuration class: [Qwen4ExpVisionModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpVisionModel) (Qwen4ExpVisionConfig model) - [RTDetrConfig](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrConfig) configuration class: [RTDetrModel](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrModel) (RTDetrConfig model) - [RTDetrV2Config](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2Config) configuration class: [RTDetrV2Model](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2Model) (RTDetrV2Config model) - [RadioConfig](/docs/transformers/main/id/model_doc/radio#transformers.RadioConfig) configuration class: [RadioModel](/docs/transformers/main/id/model_doc/radio#transformers.RadioModel) (RadioConfig model) - [RecurrentGemmaConfig](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaConfig) configuration class: [RecurrentGemmaModel](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaModel) (RecurrentGemmaConfig model) - [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig) configuration class: [ReformerModel](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerModel) (ReformerConfig model) - [RegNetConfig](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetConfig) configuration class: [RegNetModel](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetModel) (RegNetConfig model) - [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig) configuration class: [RemBertModel](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertModel) (RemBertConfig model) - [ResNetConfig](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetConfig) configuration class: [ResNetModel](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetModel) (ResNetConfig model) - [RfDetrConfig](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrConfig) configuration class: [RfDetrModel](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrModel) (RfDetrConfig model) - [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) configuration class: [RoCBertModel](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertModel) (RoCBertConfig model) - [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig) configuration class: [RoFormerModel](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerModel) (RoFormerConfig model) - [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) configuration class: [RobertaModel](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaModel) (RobertaConfig model) - [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) configuration class: [RobertaPreLayerNormModel](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormModel) (RobertaPreLayerNormConfig model) - [RwkvConfig](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvConfig) configuration class: [RwkvModel](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvModel) (RwkvConfig model) - [SEWConfig](/docs/transformers/main/id/model_doc/sew#transformers.SEWConfig) configuration class: [SEWModel](/docs/transformers/main/id/model_doc/sew#transformers.SEWModel) (SEWConfig model) - [SEWDConfig](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDConfig) configuration class: [SEWDModel](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDModel) (SEWDConfig model) - [Sam2Config](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Config) configuration class: [Sam2Model](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Model) (Sam2Config model) - [Sam2HieraDetConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2HieraDetConfig) configuration class: [Sam2HieraDetModel](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2HieraDetModel) (Sam2HieraDetConfig model) - [Sam2VideoConfig](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoConfig) configuration class: [Sam2VideoModel](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoModel) (Sam2VideoConfig model) - [Sam2VisionConfig](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2VisionConfig) configuration class: [Sam2VisionModel](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2VisionModel) (Sam2VisionConfig model) - [Sam3Config](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Config) configuration class: [Sam3Model](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Model) (Sam3Config model) - [Sam3LiteTextConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextConfig) configuration class: [Sam3LiteTextModel](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextModel) (Sam3LiteTextConfig model) - [Sam3LiteTextTextConfig](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextTextConfig) configuration class: [Sam3LiteTextTextModel](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextTextModel) (Sam3LiteTextTextConfig model) - [Sam3TrackerConfig](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerConfig) configuration class: [Sam3TrackerModel](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerModel) (Sam3TrackerConfig model) - [Sam3TrackerVideoConfig](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoConfig) configuration class: [Sam3TrackerVideoModel](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoModel) (Sam3TrackerVideoConfig model) - [Sam3ViTConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3ViTConfig) configuration class: [Sam3ViTModel](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3ViTModel) (Sam3ViTConfig model) - [Sam3VideoConfig](/docs/transformers/main/id/model_doc/sam3_video#transformers.Sam3VideoConfig) configuration class: [Sam3VideoModel](/docs/transformers/main/id/model_doc/sam3_video#transformers.Sam3VideoModel) (Sam3VideoConfig model) - [Sam3VisionConfig](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3VisionConfig) configuration class: [Sam3VisionModel](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3VisionModel) (Sam3VisionConfig model) - [SamConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamConfig) configuration class: [SamModel](/docs/transformers/main/id/model_doc/sam#transformers.SamModel) (SamConfig model) - [SamHQConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQConfig) configuration class: [SamHQModel](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQModel) (SamHQConfig model) - [SamHQVisionConfig](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQVisionConfig) configuration class: [SamHQVisionModel](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQVisionModel) (SamHQVisionConfig model) - [SamVisionConfig](/docs/transformers/main/id/model_doc/sam#transformers.SamVisionConfig) configuration class: [SamVisionModel](/docs/transformers/main/id/model_doc/sam#transformers.SamVisionModel) (SamVisionConfig model) - [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config) configuration class: [Sapiens2Model](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Model) (Sapiens2Config model) - [SeamlessM4TConfig](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TConfig) configuration class: [SeamlessM4TModel](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TModel) (SeamlessM4TConfig model) - [SeamlessM4Tv2Config](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Config) configuration class: [SeamlessM4Tv2Model](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Model) (SeamlessM4Tv2Config model) - [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig) configuration class: [SeedOssModel](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssModel) (SeedOssConfig model) - [SegGptConfig](/docs/transformers/main/id/model_doc/seggpt#transformers.SegGptConfig) configuration class: [SegGptModel](/docs/transformers/main/id/model_doc/seggpt#transformers.SegGptModel) (SegGptConfig model) - [SegformerConfig](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerConfig) configuration class: [SegformerModel](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerModel) (SegformerConfig model) - [Siglip2Config](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Config) configuration class: [Siglip2Model](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Model) (Siglip2Config model) - [Siglip2VisionConfig](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2VisionConfig) configuration class: [Siglip2VisionModel](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2VisionModel) (Siglip2VisionConfig model) - [SiglipConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipConfig) configuration class: [SiglipModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipModel) (SiglipConfig model) - [SiglipVisionConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipVisionConfig) configuration class: [SiglipVisionModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipVisionModel) (SiglipVisionConfig model) - [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config) configuration class: [SmolLM3Model](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Model) (SmolLM3Config model) - [SmolVLMConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMConfig) configuration class: [SmolVLMModel](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMModel) (SmolVLMConfig model) - [SmolVLMVisionConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMVisionConfig) configuration class: [SmolVLMVisionTransformer](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMVisionTransformer) (SmolVLMVisionConfig model) - [SolarOpenConfig](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenConfig) configuration class: [SolarOpenModel](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenModel) (SolarOpenConfig model) - [Speech2TextConfig](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextConfig) configuration class: [Speech2TextModel](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextModel) (Speech2TextConfig model) - [SpeechT5Config](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Config) configuration class: [SpeechT5Model](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Model) (SpeechT5Config model) - [SplinterConfig](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterConfig) configuration class: [SplinterModel](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterModel) (SplinterConfig model) - [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig) configuration class: [SqueezeBertModel](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertModel) (SqueezeBertConfig model) - [StableLmConfig](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmConfig) configuration class: [StableLmModel](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmModel) (StableLmConfig model) - [Starcoder2Config](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Config) configuration class: [Starcoder2Model](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Model) (Starcoder2Config model) - [Step3p7Config](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Config) configuration class: [Step3p7Model](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Model) (Step3p7Config model) - [SwiftFormerConfig](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerConfig) configuration class: [SwiftFormerModel](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerModel) (SwiftFormerConfig model) - [Swin2SRConfig](/docs/transformers/main/id/model_doc/swin2sr#transformers.Swin2SRConfig) configuration class: [Swin2SRModel](/docs/transformers/main/id/model_doc/swin2sr#transformers.Swin2SRModel) (Swin2SRConfig model) - [SwinConfig](/docs/transformers/main/id/model_doc/swin#transformers.SwinConfig) configuration class: [SwinModel](/docs/transformers/main/id/model_doc/swin#transformers.SwinModel) (SwinConfig model) - [Swinv2Config](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Config) configuration class: [Swinv2Model](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Model) (Swinv2Config model) - [SwitchTransformersConfig](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersConfig) configuration class: [SwitchTransformersModel](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersModel) (SwitchTransformersConfig model) - [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config) configuration class: [T5Model](/docs/transformers/main/id/model_doc/t5#transformers.T5Model) (T5Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2Model](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Model) (T5Gemma2Config model) - [T5Gemma2EncoderConfig](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2EncoderConfig) configuration class: `T5Gemma2Encoder` (T5Gemma2EncoderConfig model) - [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig) configuration class: [T5GemmaModel](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaModel) (T5GemmaConfig model) - [TableTransformerConfig](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerConfig) configuration class: [TableTransformerModel](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerModel) (TableTransformerConfig model) - [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig) configuration class: [TapasModel](/docs/transformers/main/id/model_doc/tapas#transformers.TapasModel) (TapasConfig model) - [TextNetConfig](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetConfig) configuration class: [TextNetModel](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetModel) (TextNetConfig model) - [TimeSeriesTransformerConfig](/docs/transformers/main/id/model_doc/time_series_transformer#transformers.TimeSeriesTransformerConfig) configuration class: [TimeSeriesTransformerModel](/docs/transformers/main/id/model_doc/time_series_transformer#transformers.TimeSeriesTransformerModel) (TimeSeriesTransformerConfig model) - [TimesFm2_5Config](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5Config) configuration class: [TimesFm2_5Model](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5Model) (TimesFm2_5Config model) - [TimesFmConfig](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmConfig) configuration class: [TimesFmModel](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmModel) (TimesFmConfig model) - [TimesformerConfig](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerConfig) configuration class: [TimesformerModel](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerModel) (TimesformerConfig model) - [TimmBackboneConfig](/docs/transformers/main/id/main_classes/backbones#transformers.TimmBackboneConfig) configuration class: [TimmBackbone](/docs/transformers/main/id/main_classes/backbones#transformers.TimmBackbone) (TimmBackboneConfig model) - [TimmWrapperConfig](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperConfig) configuration class: [TimmWrapperModel](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperModel) (TimmWrapperConfig model) - [Tipsv2Config](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Config) configuration class: [Tipsv2Model](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Model) (Tipsv2Config model) - [Tipsv2TextConfig](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2TextConfig) configuration class: [Tipsv2TextModel](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2TextModel) (Tipsv2TextConfig model) - [Tipsv2VisionConfig](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2VisionConfig) configuration class: [Tipsv2VisionModel](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2VisionModel) (Tipsv2VisionConfig model) - [TvpConfig](/docs/transformers/main/id/model_doc/tvp#transformers.TvpConfig) configuration class: [TvpModel](/docs/transformers/main/id/model_doc/tvp#transformers.TvpModel) (TvpConfig model) - [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config) configuration class: [UMT5Model](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Model) (UMT5Config model) - [UVDocConfig](/docs/transformers/main/id/model_doc/uvdoc#transformers.UVDocConfig) configuration class: [UVDocModel](/docs/transformers/main/id/model_doc/uvdoc#transformers.UVDocModel) (UVDocConfig model) - [UdopConfig](/docs/transformers/main/id/model_doc/udop#transformers.UdopConfig) configuration class: [UdopModel](/docs/transformers/main/id/model_doc/udop#transformers.UdopModel) (UdopConfig model) - [UltraVBertConfig](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertConfig) configuration class: [UltraVBertModel](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertModel) (UltraVBertConfig model) - [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig) configuration class: [UltrabertModel](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertModel) (UltrabertConfig model) - [UniSpeechConfig](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechConfig) configuration class: [UniSpeechModel](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechModel) (UniSpeechConfig model) - [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig) configuration class: [UniSpeechSatModel](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatModel) (UniSpeechSatConfig model) - [UnivNetConfig](/docs/transformers/main/id/model_doc/univnet#transformers.UnivNetConfig) configuration class: [UnivNetModel](/docs/transformers/main/id/model_doc/univnet#transformers.UnivNetModel) (UnivNetConfig model) - [UnlimitedOcrConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrConfig) configuration class: [UnlimitedOcrModel](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrModel) (UnlimitedOcrConfig model) - [VJEPA2Config](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2Config) configuration class: [VJEPA2Model](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2Model) (VJEPA2Config model) - [VaultGemmaConfig](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaConfig) configuration class: [VaultGemmaModel](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaModel) (VaultGemmaConfig model) - [ViTConfig](/docs/transformers/main/id/model_doc/vit#transformers.ViTConfig) configuration class: [ViTModel](/docs/transformers/main/id/model_doc/vit#transformers.ViTModel) (ViTConfig model) - [ViTMAEConfig](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEConfig) configuration class: [ViTMAEModel](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEModel) (ViTMAEConfig model) - [ViTMSNConfig](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNConfig) configuration class: [ViTMSNModel](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNModel) (ViTMSNConfig model) - [VibeVoiceAcousticTokenizerConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerConfig) configuration class: [VibeVoiceAcousticTokenizerModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerModel) (VibeVoiceAcousticTokenizerConfig model) - [VibeVoiceAcousticTokenizerDecoderConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerDecoderConfig) configuration class: [VibeVoiceAcousticTokenizerDecoderModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerDecoderModel) (VibeVoiceAcousticTokenizerDecoderConfig model) - [VibeVoiceAcousticTokenizerEncoderConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerEncoderConfig) configuration class: [VibeVoiceAcousticTokenizerEncoderModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerEncoderModel) (VibeVoiceAcousticTokenizerEncoderConfig model) - [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig) configuration class: [VibeVoiceAsrModel](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrModel) (VibeVoiceAsrConfig model) - [VibeVoiceConfig](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceConfig) configuration class: [VibeVoiceForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceForConditionalGeneration) (VibeVoiceConfig model) - [VideoLlama3Config](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Config) configuration class: [VideoLlama3Model](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Model) (VideoLlama3Config model) - [VideoLlama3VisionConfig](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3VisionConfig) configuration class: [VideoLlama3VisionModel](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3VisionModel) (VideoLlama3VisionConfig model) - [VideoLlavaConfig](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaConfig) configuration class: [VideoLlavaModel](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaModel) (VideoLlavaConfig model) - [VideoMAEConfig](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEConfig) configuration class: [VideoMAEModel](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEModel) (VideoMAEConfig model) - [VideoPrismConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismConfig) configuration class: [VideoPrismClipModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismClipModel) (VideoPrismConfig model) - [VideoPrismVisionConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismVisionConfig) configuration class: [VideoPrismVisionModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismVisionModel) (VideoPrismVisionConfig model) - [ViltConfig](/docs/transformers/main/id/model_doc/vilt#transformers.ViltConfig) configuration class: [ViltModel](/docs/transformers/main/id/model_doc/vilt#transformers.ViltModel) (ViltConfig model) - [VipLlavaConfig](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaConfig) configuration class: [VipLlavaModel](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaModel) (VipLlavaConfig model) - [VisionTextDualEncoderConfig](/docs/transformers/main/id/model_doc/vision-text-dual-encoder#transformers.VisionTextDualEncoderConfig) configuration class: [VisionTextDualEncoderModel](/docs/transformers/main/id/model_doc/vision-text-dual-encoder#transformers.VisionTextDualEncoderModel) (VisionTextDualEncoderConfig model) - [VisualBertConfig](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertConfig) configuration class: [VisualBertModel](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertModel) (VisualBertConfig model) - [VitDetConfig](/docs/transformers/main/id/model_doc/vitdet#transformers.VitDetConfig) configuration class: [VitDetModel](/docs/transformers/main/id/model_doc/vitdet#transformers.VitDetModel) (VitDetConfig model) - [VitsConfig](/docs/transformers/main/id/model_doc/vits#transformers.VitsConfig) configuration class: [VitsModel](/docs/transformers/main/id/model_doc/vits#transformers.VitsModel) (VitsConfig model) - [VivitConfig](/docs/transformers/main/id/model_doc/vivit#transformers.VivitConfig) configuration class: [VivitModel](/docs/transformers/main/id/model_doc/vivit#transformers.VivitModel) (VivitConfig model) - [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig) configuration class: [VoxtralModel](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralModel) (VoxtralConfig model) - [VoxtralEncoderConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralEncoderConfig) configuration class: [VoxtralEncoder](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralEncoder) (VoxtralEncoderConfig model) - [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig) configuration class: [VoxtralRealtimeModel](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeModel) (VoxtralRealtimeConfig model) - [VoxtralRealtimeEncoderConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeEncoderConfig) configuration class: [VoxtralRealtimeEncoder](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeEncoder) (VoxtralRealtimeEncoderConfig model) - [VoxtralRealtimeTextConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeTextConfig) configuration class: `VoxtralRealtimeTextModel` (VoxtralRealtimeTextConfig model) - [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig) configuration class: [Wav2Vec2BertModel](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertModel) (Wav2Vec2BertConfig model) - [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config) configuration class: [Wav2Vec2Model](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Model) (Wav2Vec2Config model) - [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig) configuration class: [Wav2Vec2ConformerModel](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerModel) (Wav2Vec2ConformerConfig model) - [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig) configuration class: [WavLMModel](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMModel) (WavLMConfig model) - [WhisperConfig](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperConfig) configuration class: [WhisperModel](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperModel) (WhisperConfig model) - [XCLIPConfig](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPConfig) configuration class: [XCLIPModel](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPModel) (XCLIPConfig model) - [XGLMConfig](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMConfig) configuration class: [XGLMModel](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMModel) (XGLMConfig model) - [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) configuration class: [XLMModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMModel) (XLMConfig model) - [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) configuration class: [XLMRobertaModel](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaModel) (XLMRobertaConfig model) - [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) configuration class: [XLMRobertaXLModel](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLModel) (XLMRobertaXLConfig model) - [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) configuration class: [XLNetModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetModel) (XLNetConfig model) - [Xcodec2Config](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Config) configuration class: [Xcodec2Model](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Model) (Xcodec2Config model) - [XcodecConfig](/docs/transformers/main/id/model_doc/xcodec#transformers.XcodecConfig) configuration class: [XcodecModel](/docs/transformers/main/id/model_doc/xcodec#transformers.XcodecModel) (XcodecConfig model) - [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) configuration class: [XmodModel](/docs/transformers/main/id/model_doc/xmod#transformers.XmodModel) (XmodConfig model) - [YolosConfig](/docs/transformers/main/id/model_doc/yolos#transformers.YolosConfig) configuration class: [YolosModel](/docs/transformers/main/id/model_doc/yolos#transformers.YolosModel) (YolosConfig model) - [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig) configuration class: [YosoModel](/docs/transformers/main/id/model_doc/yoso#transformers.YosoModel) (YosoConfig model) - [YoutuConfig](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuConfig) configuration class: [YoutuModel](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuModel) (YoutuConfig model) - [Zamba2Config](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Config) configuration class: [Zamba2Model](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Model) (Zamba2Config model) - [ZambaConfig](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaConfig) configuration class: [ZambaModel](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaModel) (ZambaConfig model) - [ZayaConfig](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaConfig) configuration class: [ZayaModel](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaModel) (ZayaConfig model) - [xLSTMConfig](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMConfig) configuration class: [xLSTMModel](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMModel) (xLSTMConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dasar pustaka dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModel

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModel.from_config(config)
```

#### from_pretrained[[transformers.AutoModel.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dasar dari pustaka ini dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **afmoe** -- [AfmoeModel](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeModel) (model AfmoeConfig)
- **aimv2** -- [Aimv2Model](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2Model) (model Aimv2Config)
- **aimv2_vision_model** -- [Aimv2VisionModel](/docs/transformers/main/id/model_doc/aimv2#transformers.Aimv2VisionModel) (model Aimv2VisionConfig)
- **albert** -- [AlbertModel](/docs/transformers/main/id/model_doc/albert#transformers.AlbertModel) (model AlbertConfig)
- **align** -- [AlignModel](/docs/transformers/main/id/model_doc/align#transformers.AlignModel) (model AlignConfig)
- **altclip** -- [AltCLIPModel](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPModel) (model AltCLIPConfig)
- **apertus** -- [ApertusModel](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusModel) (model ApertusConfig)
- **arcee** -- [ArceeModel](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeModel) (model ArceeConfig)
- **aria** -- [AriaModel](/docs/transformers/main/id/model_doc/aria#transformers.AriaModel) (model AriaConfig)
- **aria_text** -- [AriaTextModel](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextModel) (model AriaTextConfig)
- **audio-spectrogram-transformer** -- [ASTModel](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTModel) (model ASTConfig)
- **audioflamingo3** -- [AudioFlamingo3Model](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Model) (model AudioFlamingo3Config)
- **audioflamingo3_encoder** -- [AudioFlamingo3Encoder](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Encoder) (model AudioFlamingo3EncoderConfig)
- **autoformer** -- [AutoformerModel](/docs/transformers/main/id/model_doc/autoformer#transformers.AutoformerModel) (model AutoformerConfig)
- **axk1** -- [AXK1Model](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Model) (model AXK1Config)
- **axk2** -- [AXK2Model](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Model) (model AXK2Config)
- **aya_vision** -- [AyaVisionModel](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionModel) (model AyaVisionConfig)
- **bailing_hybrid** -- [BailingHybridModel](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridModel) (model BailingHybridConfig)
- **bamba** -- [BambaModel](/docs/transformers/main/id/model_doc/bamba#transformers.BambaModel) (model BambaConfig)
- **bark** -- [BarkModel](/docs/transformers/main/id/model_doc/bark#transformers.BarkModel) (model BarkConfig)
- **bart** -- [BartModel](/docs/transformers/main/id/model_doc/bart#transformers.BartModel) (model BartConfig)
- **beam** -- [BeamModel](/docs/transformers/main/id/model_doc/beam#transformers.BeamModel) (model BeamConfig)
- **beit** -- [BeitModel](/docs/transformers/main/id/model_doc/beit#transformers.BeitModel) (model BeitConfig)
- **bert** -- [BertModel](/docs/transformers/main/id/model_doc/bert#transformers.BertModel) (model BertConfig)
- **bert-generation** -- [BertGenerationEncoder](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationEncoder) (model BertGenerationConfig)
- **big_bird** -- [BigBirdModel](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdModel) (model BigBirdConfig)
- **bigbird_pegasus** -- [BigBirdPegasusModel](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusModel) (model BigBirdPegasusConfig)
- **biogpt** -- [BioGptModel](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptModel) (model BioGptConfig)
- **bit** -- [BitModel](/docs/transformers/main/id/model_doc/bit#transformers.BitModel) (model BitConfig)
- **bitnet** -- [BitNetModel](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetModel) (model BitNetConfig)
- **blenderbot** -- [BlenderbotModel](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotModel) (model BlenderbotConfig)
- **blenderbot-small** -- [BlenderbotSmallModel](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallModel) (model BlenderbotSmallConfig)
- **blip** -- [BlipModel](/docs/transformers/main/id/model_doc/blip#transformers.BlipModel) (model BlipConfig)
- **blip-2** -- [Blip2Model](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Model) (model Blip2Config)
- **blip_2_qformer** -- [Blip2QFormerModel](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2QFormerModel) (model Blip2QFormerConfig)
- **bloom** -- [BloomModel](/docs/transformers/main/id/model_doc/bloom#transformers.BloomModel) (model BloomConfig)
- **blt** -- [BltModel](/docs/transformers/main/id/model_doc/blt#transformers.BltModel) (model BltConfig)
- **bridgetower** -- [BridgeTowerModel](/docs/transformers/main/id/model_doc/bridgetower#transformers.BridgeTowerModel) (model BridgeTowerConfig)
- **bros** -- [BrosModel](/docs/transformers/main/id/model_doc/bros#transformers.BrosModel) (model BrosConfig)
- **camembert** -- [CamembertModel](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertModel) (model CamembertConfig)
- **canary** -- [CanaryModel](/docs/transformers/main/id/model_doc/canary#transformers.CanaryModel) (model CanaryConfig)
- **canine** -- [CanineModel](/docs/transformers/main/id/model_doc/canine#transformers.CanineModel) (model CanineConfig)
- **chameleon** -- [ChameleonModel](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonModel) (model ChameleonConfig)
- **chinese_clip** -- [ChineseCLIPModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPModel) (model ChineseCLIPConfig)
- **chinese_clip_vision_model** -- [ChineseCLIPVisionModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPVisionModel) (model ChineseCLIPVisionConfig)
- **clap** -- [ClapModel](/docs/transformers/main/id/model_doc/clap#transformers.ClapModel) (model ClapConfig)
- **clip** -- [CLIPModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPModel) (model CLIPConfig)
- **clip_text_model** -- [CLIPTextModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPTextModel) (model CLIPTextConfig)
- **clip_vision_model** -- [CLIPVisionModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPVisionModel) (model CLIPVisionConfig)
- **clipseg** -- [CLIPSegModel](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegModel) (model CLIPSegConfig)
- **clvp** -- [ClvpModelForConditionalGeneration](/docs/transformers/main/id/model_doc/clvp#transformers.ClvpModelForConditionalGeneration) (model ClvpConfig)
- **codegen** -- [CodeGenModel](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenModel) (model CodeGenConfig)
- **cohere** -- [CohereModel](/docs/transformers/main/id/model_doc/cohere#transformers.CohereModel) (model CohereConfig)
- **cohere2** -- [Cohere2Model](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2Model) (model Cohere2Config)
- **cohere2_moe** -- [Cohere2MoeModel](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeModel) (model Cohere2MoeConfig)
- **cohere2_vision** -- [Cohere2VisionModel](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionModel) (model Cohere2VisionConfig)
- **cohere_asr** -- [CohereAsrModel](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrModel) (model CohereAsrConfig)
- **cohere_compass** -- [CohereCompassModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassModel) (model CohereCompassConfig)
- **cohere_compass_text** -- [CohereCompassTextModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextModel) (model CohereCompassTextConfig)
- **cohere_compass_vision** -- [CohereCompassVisionModel](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassVisionModel) (model CohereCompassVisionConfig)
- **conditional_detr** -- [ConditionalDetrModel](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrModel) (model ConditionalDetrConfig)
- **convbert** -- [ConvBertModel](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertModel) (model ConvBertConfig)
- **convnext** -- [ConvNextModel](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextModel) (model ConvNextConfig)
- **convnextv2** -- [ConvNextV2Model](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2Model) (model ConvNextV2Config)
- **cosmos3_edge** -- [Cosmos3EdgeModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeModel) (model Cosmos3EdgeConfig)
- **cosmos3_edge_text** -- [Cosmos3EdgeTextModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeTextModel) (model Cosmos3EdgeTextConfig)
- **cosmos3_edge_vision** -- [Cosmos3EdgeVisionModel](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeVisionModel) (model Cosmos3EdgeVisionConfig)
- **cosmos3_omni** -- [Cosmos3OmniModel](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniModel) (model Cosmos3OmniConfig)
- **cpmant** -- [CpmAntModel](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntModel) (model CpmAntConfig)
- **csm** -- [CsmForConditionalGeneration](/docs/transformers/main/id/model_doc/csm#transformers.CsmForConditionalGeneration) (model CsmConfig)
- **ctrl** -- [CTRLModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLModel) (model CTRLConfig)
- **cvt** -- [CvtModel](/docs/transformers/main/id/model_doc/cvt#transformers.CvtModel) (model CvtConfig)
- **cwm** -- [CwmModel](/docs/transformers/main/id/model_doc/cwm#transformers.CwmModel) (model CwmConfig)
- **d_fine** -- [DFineModel](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineModel) (model DFineConfig)
- **dab-detr** -- [DabDetrModel](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrModel) (model DabDetrConfig)
- **dac** -- [DacModel](/docs/transformers/main/id/model_doc/dac#transformers.DacModel) (model DacConfig)
- **data2vec-audio** -- [Data2VecAudioModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioModel) (model Data2VecAudioConfig)
- **data2vec-text** -- [Data2VecTextModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextModel) (model Data2VecTextConfig)
- **data2vec-vision** -- [Data2VecVisionModel](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionModel) (model Data2VecVisionConfig)
- **dbrx** -- [DbrxModel](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxModel) (model DbrxConfig)
- **deberta** -- [DebertaModel](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaModel) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2Model](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Model) (model DebertaV2Config)
- **decision_transformer** -- [DecisionTransformerModel](/docs/transformers/main/id/model_doc/decision_transformer#transformers.DecisionTransformerModel) (model DecisionTransformerConfig)
- **deepseek_ocr2** -- [DeepseekOcr2Model](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Model) (model DeepseekOcr2Config)
- **deepseek_v2** -- [DeepseekV2Model](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Model) (model DeepseekV2Config)
- **deepseek_v3** -- [DeepseekV3Model](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Model) (model DeepseekV3Config)
- **deepseek_v32** -- [DeepseekV32Model](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32Model) (model DeepseekV32Config)
- **deepseek_v4** -- [DeepseekV4Model](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4Model) (model DeepseekV4Config)
- **deepseek_vl** -- [DeepseekVLModel](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLModel) (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- [DeepseekVLHybridModel](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridModel) (model DeepseekVLHybridConfig)
- **deformable_detr** -- [DeformableDetrModel](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrModel) (model DeformableDetrConfig)
- **deimv2** -- [Deimv2Model](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2Model) (model Deimv2Config)
- **deit** -- [DeiTModel](/docs/transformers/main/id/model_doc/deit#transformers.DeiTModel) (model DeiTConfig)
- **depth_pro** -- [DepthProModel](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProModel) (model DepthProConfig)
- **detr** -- [DetrModel](/docs/transformers/main/id/model_doc/detr#transformers.DetrModel) (model DetrConfig)
- **dia** -- [DiaModel](/docs/transformers/main/id/model_doc/dia#transformers.DiaModel) (model DiaConfig)
- **diffllama** -- [DiffLlamaModel](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaModel) (model DiffLlamaConfig)
- **diffusion_gemma** -- [DiffusionGemmaModel](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaModel) (model DiffusionGemmaConfig)
- **dinat** -- [DinatModel](/docs/transformers/main/id/model_doc/dinat#transformers.DinatModel) (model DinatConfig)
- **dinov2** -- [Dinov2Model](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2Model) (model Dinov2Config)
- **dinov2_with_registers** -- [Dinov2WithRegistersModel](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersModel) (model Dinov2WithRegistersConfig)
- **dinov3_convnext** -- [DINOv3ConvNextModel](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ConvNextModel) (model DINOv3ConvNextConfig)
- **dinov3_vit** -- [DINOv3ViTModel](/docs/transformers/main/id/model_doc/dinov3#transformers.DINOv3ViTModel) (model DINOv3ViTConfig)
- **distilbert** -- [DistilBertModel](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertModel) (model DistilBertConfig)
- **doge** -- [DogeModel](/docs/transformers/main/id/model_doc/doge#transformers.DogeModel) (model DogeConfig)
- **donut-swin** -- [DonutSwinModel](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinModel) (model DonutSwinConfig)
- **dots1** -- [Dots1Model](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1Model) (model Dots1Config)
- **dpr** -- [DPRQuestionEncoder](/docs/transformers/main/id/model_doc/dpr#transformers.DPRQuestionEncoder) (model DPRConfig)
- **dpt** -- [DPTModel](/docs/transformers/main/id/model_doc/dpt#transformers.DPTModel) (model DPTConfig)
- **edgetam** -- [EdgeTamModel](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamModel) (model EdgeTamConfig)
- **edgetam_video** -- [EdgeTamVideoModel](/docs/transformers/main/id/model_doc/edgetam_video#transformers.EdgeTamVideoModel) (model EdgeTamVideoConfig)
- **edgetam_vision_model** -- [EdgeTamVisionModel](/docs/transformers/main/id/model_doc/edgetam#transformers.EdgeTamVisionModel) (model EdgeTamVisionConfig)
- **efficientloftr** -- [EfficientLoFTRModel](/docs/transformers/main/id/model_doc/efficientloftr#transformers.EfficientLoFTRModel) (model EfficientLoFTRConfig)
- **efficientnet** -- [EfficientNetModel](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetModel) (model EfficientNetConfig)
- **electra** -- [ElectraModel](/docs/transformers/main/id/model_doc/electra#transformers.ElectraModel) (model ElectraConfig)
- **embedding_gemma2** -- [EmbeddingGemma2Model](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2Model) (model EmbeddingGemma2Config)
- **embedding_gemma2_text** -- [EmbeddingGemma2TextModel](/docs/transformers/main/id/model_doc/embedding_gemma2#transformers.EmbeddingGemma2TextModel) (model EmbeddingGemma2TextConfig)
- **emu3** -- [Emu3Model](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Model) (model Emu3Config)
- **encodec** -- [EncodecModel](/docs/transformers/main/id/model_doc/encodec#transformers.EncodecModel) (model EncodecConfig)
- **ernie** -- [ErnieModel](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieModel) (model ErnieConfig)
- **ernie4_5** -- [Ernie4_5Model](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5Model) (model Ernie4_5Config)
- **ernie4_5_moe** -- [Ernie4_5_MoeModel](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeModel) (model Ernie4_5_MoeConfig)
- **ernie4_5_vl_moe** -- [Ernie4_5_VLMoeModel](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeModel) (model Ernie4_5_VLMoeConfig)
- **esm** -- [EsmModel](/docs/transformers/main/id/model_doc/esm#transformers.EsmModel) (model EsmConfig)
- **esmc** -- [EsmcModel](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcModel) (model EsmcConfig)
- **eurobert** -- [EuroBertModel](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertModel) (model EuroBertConfig)
- **evolla** -- [EvollaModel](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaModel) (model EvollaConfig)
- **exaone4** -- [Exaone4Model](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Model) (model Exaone4Config)
- **exaone4_5** -- [Exaone4_5_Model](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Model) (model Exaone4_5_Config)
- **exaone4_5_vision** -- [Exaone4_5_VisionModel](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_VisionModel) (model Exaone4_5_VisionConfig)
- **exaone_moe** -- [ExaoneMoeModel](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeModel) (model ExaoneMoeConfig)
- **falcon** -- [FalconModel](/docs/transformers/main/id/model_doc/falcon#transformers.FalconModel) (model FalconConfig)
- **falcon_h1** -- [FalconH1Model](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1Model) (model FalconH1Config)
- **falcon_mamba** -- [FalconMambaModel](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaModel) (model FalconMambaConfig)
- **fast_vlm** -- [FastVlmModel](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmModel) (model FastVlmConfig)
- **fastspeech2_conformer** -- [FastSpeech2ConformerModel](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerModel) (model FastSpeech2ConformerConfig)
- **fastspeech2_conformer_with_hifigan** -- [FastSpeech2ConformerWithHifiGan](/docs/transformers/main/id/model_doc/fastspeech2_conformer#transformers.FastSpeech2ConformerWithHifiGan) (model FastSpeech2ConformerWithHifiGanConfig)
- **flaubert** -- [FlaubertModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertModel) (model FlaubertConfig)
- **flava** -- [FlavaModel](/docs/transformers/main/id/model_doc/flava#transformers.FlavaModel) (model FlavaConfig)
- **flex_olmo** -- [FlexOlmoModel](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoModel) (model FlexOlmoConfig)
- **flint** -- [FlintModel](/docs/transformers/main/id/model_doc/flint#transformers.FlintModel) (model FlintConfig)
- **florence2** -- [Florence2Model](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Model) (model Florence2Config)
- **fnet** -- [FNetModel](/docs/transformers/main/id/model_doc/fnet#transformers.FNetModel) (model FNetConfig)
- **focalnet** -- [FocalNetModel](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetModel) (model FocalNetConfig)
- **fsmt** -- [FSMTModel](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTModel) (model FSMTConfig)
- **fun_asr_nano** -- [FunAsrNanoModel](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoModel) (model FunAsrNanoConfig)
- **fun_asr_nano_encoder** -- [FunAsrNanoEncoder](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoEncoder) (model FunAsrNanoEncoderConfig)
- **funnel** -- [FunnelModel](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelModel) atau [FunnelBaseModel](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelBaseModel) (model FunnelConfig)
- **fuyu** -- [FuyuModel](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuModel) (model FuyuConfig)
- **gemma** -- [GemmaModel](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaModel) (model GemmaConfig)
- **gemma2** -- [Gemma2Model](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Model) (model Gemma2Config)
- **gemma3** -- [Gemma3Model](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Model) (model Gemma3Config)
- **gemma3_text** -- [Gemma3TextModel](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextModel) (model Gemma3TextConfig)
- **gemma3n** -- [Gemma3nModel](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nModel) (model Gemma3nConfig)
- **gemma3n_audio** -- `Gemma3nAudioEncoder` (model Gemma3nAudioConfig)
- **gemma3n_text** -- [Gemma3nTextModel](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nTextModel) (model Gemma3nTextConfig)
- **gemma3n_vision** -- [TimmWrapperModel](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperModel) (model Gemma3nVisionConfig)
- **gemma4** -- [Gemma4Model](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Model) (model Gemma4Config)
- **gemma4_audio** -- [Gemma4AudioModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4AudioModel) (model Gemma4AudioConfig)
- **gemma4_text** -- [Gemma4TextModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4TextModel) (model Gemma4TextConfig)
- **gemma4_unified** -- [Gemma4UnifiedModel](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedModel) (model Gemma4UnifiedConfig)
- **gemma4_unified_text** -- [Gemma4UnifiedTextModel](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedTextModel) (model Gemma4UnifiedTextConfig)
- **gemma4_vision** -- [Gemma4VisionModel](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4VisionModel) (model Gemma4VisionConfig)
- **git** -- [GitModel](/docs/transformers/main/id/model_doc/git#transformers.GitModel) (model GitConfig)
- **glm** -- [GlmModel](/docs/transformers/main/id/model_doc/glm#transformers.GlmModel) (model GlmConfig)
- **glm4** -- [Glm4Model](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Model) (model Glm4Config)
- **glm46v** -- [Glm46VModel](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VModel) (model Glm46VConfig)
- **glm4_moe** -- [Glm4MoeModel](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeModel) (model Glm4MoeConfig)
- **glm4_moe_lite** -- [Glm4MoeLiteModel](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteModel) (model Glm4MoeLiteConfig)
- **glm4v** -- [Glm4vModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vModel) (model Glm4vConfig)
- **glm4v_moe** -- [Glm4vMoeModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeModel) (model Glm4vMoeConfig)
- **glm4v_moe_text** -- [Glm4vMoeTextModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeTextModel) (model Glm4vMoeTextConfig)
- **glm4v_moe_vision** -- [Glm4vMoeVisionModel](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeVisionModel) (model Glm4vMoeVisionConfig)
- **glm4v_text** -- [Glm4vTextModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vTextModel) (model Glm4vTextConfig)
- **glm4v_vision** -- [Glm4vVisionModel](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vVisionModel) (model Glm4vVisionConfig)
- **glm5_next** -- [Glm5NextModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextModel) (model Glm5NextConfig)
- **glm5_next_text** -- [Glm5NextTextModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextTextModel) (model Glm5NextTextConfig)
- **glm5_next_vision** -- [Glm5NextVisionModel](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextVisionModel) (model Glm5NextVisionConfig)
- **glm_image** -- [GlmImageModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageModel) (model GlmImageConfig)
- **glm_image_text** -- [GlmImageTextModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageTextModel) (model GlmImageTextConfig)
- **glm_image_vision** -- [GlmImageVisionModel](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVisionModel) (model GlmImageVisionConfig)
- **glm_image_vqmodel** -- [GlmImageVQVAE](/docs/transformers/main/id/model_doc/glm_image#transformers.GlmImageVQVAE) (model GlmImageVQVAEConfig)
- **glm_moe_dsa** -- [GlmMoeDsaModel](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaModel) (model GlmMoeDsaConfig)
- **glm_ocr** -- [GlmOcrModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrModel) (model GlmOcrConfig)
- **glm_ocr_text** -- [GlmOcrTextModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrTextModel) (model GlmOcrTextConfig)
- **glm_ocr_vision** -- [GlmOcrVisionModel](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrVisionModel) (model GlmOcrVisionConfig)
- **glmasr** -- [GlmAsrModel](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrModel) (model GlmAsrConfig)
- **glmasr_encoder** -- [GlmAsrEncoder](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrEncoder) (model GlmAsrEncoderConfig)
- **glmga** -- [Glm46VModel](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VModel) (model GlmgaConfig)
- **glpn** -- [GLPNModel](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNModel) (model GLPNConfig)
- **got_ocr2** -- [GotOcr2Model](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Model) (model GotOcr2Config)
- **gpt-sw3** -- [GPT2Model](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Model) (model GPT2Config)
- **gpt2** -- [GPT2Model](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Model) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeModel](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeModel) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPTNeoModel](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoModel) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXModel](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXModel) (model GPTNeoXConfig)
- **gpt_neox_japanese** -- [GPTNeoXJapaneseModel](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseModel) (model GPTNeoXJapaneseConfig)
- **gpt_oss** -- [GptOssModel](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssModel) (model GptOssConfig)
- **gptj** -- [GPTJModel](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJModel) (model GPTJConfig)
- **gradium_phonon_tts** -- [GradiumPhononTTSModel](/docs/transformers/main/id/model_doc/gradium_phonon_tts#transformers.GradiumPhononTTSModel) (model GradiumPhononTTSConfig)
- **granite** -- [GraniteModel](/docs/transformers/main/id/model_doc/granite#transformers.GraniteModel) (model GraniteConfig)
- **granite4_vision** -- [Granite4VisionModel](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionModel) (model Granite4VisionConfig)
- **granite_speech** -- [GraniteSpeechModel](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechModel) (model GraniteSpeechConfig)
- **granite_speech5_ctc** -- [GraniteSpeech5ForCTC](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5ForCTC) (model GraniteSpeech5CTCConfig)
- **granite_speech5_encoder** -- [GraniteSpeech5Encoder](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5Encoder) (model GraniteSpeech5EncoderConfig)
- **granite_speech_plus** -- [GraniteSpeechPlusModel](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusModel) (model GraniteSpeechPlusConfig)
- **granite_swa** -- [GraniteSWAModel](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAModel) (model GraniteSWAConfig)
- **granitemoe** -- [GraniteMoeModel](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeModel) (model GraniteMoeConfig)
- **granitemoe_swa** -- [GraniteMoeSWAModel](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAModel) (model GraniteMoeSWAConfig)
- **granitemoehybrid** -- [GraniteMoeHybridModel](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridModel) (model GraniteMoeHybridConfig)
- **granitemoeshared** -- [GraniteMoeSharedModel](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedModel) (model GraniteMoeSharedConfig)
- **grounding-dino** -- [GroundingDinoModel](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoModel) (model GroundingDinoConfig)
- **groupvit** -- [GroupViTModel](/docs/transformers/main/id/model_doc/groupvit#transformers.GroupViTModel) (model GroupViTConfig)
- **gte** -- [GteModel](/docs/transformers/main/id/model_doc/gte#transformers.GteModel) (model GteConfig)
- **helium** -- [HeliumModel](/docs/transformers/main/id/model_doc/helium#transformers.HeliumModel) (model HeliumConfig)
- **hgnet_v2** -- [HGNetV2Backbone](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2Backbone) (model HGNetV2Config)
- **hiera** -- [HieraModel](/docs/transformers/main/id/model_doc/hiera#transformers.HieraModel) (model HieraConfig)
- **higgs_audio_v2** -- [HiggsAudioV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/higgs_audio_v2#transformers.HiggsAudioV2ForConditionalGeneration) (model HiggsAudioV2Config)
- **higgs_audio_v2_tokenizer** -- [HiggsAudioV2TokenizerModel](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerModel) (model HiggsAudioV2TokenizerConfig)
- **hrm_text** -- [HrmTextModel](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextModel) (model HrmTextConfig)
- **hubert** -- [HubertModel](/docs/transformers/main/id/model_doc/hubert#transformers.HubertModel) (model HubertConfig)
- **hunyuan_v1_dense** -- [HunYuanDenseV1Model](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Model) (model HunYuanDenseV1Config)
- **hunyuan_v1_moe** -- [HunYuanMoEV1Model](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Model) (model HunYuanMoEV1Config)
- **hunyuan_vl** -- [HunYuanVLModel](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLModel) (model HunYuanVLConfig)
- **hunyuan_vl_text** -- [HunYuanVLTextModel](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLTextModel) (model HunYuanVLTextConfig)
- **hunyuan_vl_vision** -- [HunYuanVLVisionTransformer](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLVisionTransformer) (model HunYuanVLVisionConfig)
- **hy_v3** -- [HYV3Model](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3Model) (model HYV3Config)
- **hy_v4** -- [HYV4Model](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4Model) (model HYV4Config)
- **hyperclovax** -- [HyperCLOVAXModel](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXModel) (model HyperCLOVAXConfig)
- **hyperclovax_vision_v2** -- [HyperCLOVAXVisionV2Model](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Model) (model HyperCLOVAXVisionV2Config)
- **ibert** -- [IBertModel](/docs/transformers/main/id/model_doc/ibert#transformers.IBertModel) (model IBertConfig)
- **idefics** -- [IdeficsModel](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsModel) (model IdeficsConfig)
- **idefics2** -- [Idefics2Model](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Model) (model Idefics2Config)
- **idefics3** -- [Idefics3Model](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Model) (model Idefics3Config)
- **idefics3_vision** -- [Idefics3VisionTransformer](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3VisionTransformer) (model Idefics3VisionConfig)
- **ijepa** -- [IJepaModel](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaModel) (model IJepaConfig)
- **imagegpt** -- [ImageGPTModel](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTModel) (model ImageGPTConfig)
- **informer** -- [InformerModel](/docs/transformers/main/id/model_doc/informer#transformers.InformerModel) (model InformerConfig)
- **inkling_mm_model** -- [InklingModel](/docs/transformers/main/id/model_doc/inkling#transformers.InklingModel) (model InklingConfig)
- **inkling_text** -- [InklingTextModel](/docs/transformers/main/id/model_doc/inkling#transformers.InklingTextModel) (model InklingTextConfig)
- **instructblip** -- [InstructBlipModel](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipModel) (model InstructBlipConfig)
- **instructblipvideo** -- [InstructBlipVideoModel](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoModel) (model InstructBlipVideoConfig)
- **internvl** -- [InternVLModel](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLModel) (model InternVLConfig)
- **internvl_vision** -- [InternVLVisionModel](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLVisionModel) (model InternVLVisionConfig)
- **jais2** -- [Jais2Model](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2Model) (model Jais2Config)
- **jamba** -- [JambaModel](/docs/transformers/main/id/model_doc/jamba#transformers.JambaModel) (model JambaConfig)
- **janus** -- [JanusModel](/docs/transformers/main/id/model_doc/janus#transformers.JanusModel) (model JanusConfig)
- **jetmoe** -- [JetMoeModel](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeModel) (model JetMoeConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3Model](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Model) (model JinaEmbeddingsV3Config)
- **kimi_k25** -- [Kimi_K25Model](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Model) (model Kimi_K25Config)
- **kimi_k25_vision** -- [Kimi_K25VisionModel](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25VisionModel) (model Kimi_K25VisionConfig)
- **kimi_linear** -- [KimiLinearModel](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearModel) (model KimiLinearConfig)
- **kolibri1** -- [Kolibri1Model](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1Model) (model Kolibri1Config)
- **kosmos-2** -- [Kosmos2Model](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Model) (model Kosmos2Config)
- **kosmos-2.5** -- [Kosmos2_5Model](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Model) (model Kosmos2_5Config)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextModel](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextModel) (model KyutaiSpeechToTextConfig)
- **laguna** -- [LagunaModel](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaModel) (model LagunaConfig)
- **lasr_ctc** -- [LasrForCTC](/docs/transformers/main/id/model_doc/lasr#transformers.LasrForCTC) (model LasrCTCConfig)
- **lasr_encoder** -- [LasrEncoder](/docs/transformers/main/id/model_doc/lasr#transformers.LasrEncoder) (model LasrEncoderConfig)
- **layoutlm** -- [LayoutLMModel](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMModel) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2Model](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Model) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3Model](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Model) (model LayoutLMv3Config)
- **led** -- [LEDModel](/docs/transformers/main/id/model_doc/led#transformers.LEDModel) (model LEDConfig)
- **levit** -- [LevitModel](/docs/transformers/main/id/model_doc/levit#transformers.LevitModel) (model LevitConfig)
- **lfm2** -- [Lfm2Model](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2Model) (model Lfm2Config)
- **lfm2_moe** -- [Lfm2MoeModel](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeModel) (model Lfm2MoeConfig)
- **lfm2_vl** -- [Lfm2VlModel](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlModel) (model Lfm2VlConfig)
- **lfm3** -- [Lfm3Model](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3Model) (model Lfm3Config)
- **lightglue** -- [LightGlueForKeypointMatching](/docs/transformers/main/id/model_doc/lightglue#transformers.LightGlueForKeypointMatching) (model LightGlueConfig)
- **lighton_ocr** -- [LightOnOcrModel](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrModel) (model LightOnOcrConfig)
- **lilt** -- [LiltModel](/docs/transformers/main/id/model_doc/lilt#transformers.LiltModel) (model LiltConfig)
- **llama** -- [LlamaModel](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaModel) (model LlamaConfig)
- **llama4** -- [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (model Llama4Config)
- **llama4_text** -- [Llama4TextModel](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4TextModel) (model Llama4TextConfig)
- **llava** -- [LlavaModel](/docs/transformers/main/id/model_doc/llava#transformers.LlavaModel) (model LlavaConfig)
- **llava_next** -- [LlavaNextModel](/docs/transformers/main/id/model_doc/llava_next#transformers.LlavaNextModel) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoModel](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoModel) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionModel](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionModel) (model LlavaOnevisionConfig)
- **longcat_flash** -- [LongcatFlashModel](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashModel) (model LongcatFlashConfig)
- **longformer** -- [LongformerModel](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerModel) (model LongformerConfig)
- **longt5** -- [LongT5Model](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5Model) (model LongT5Config)
- **luke** -- [LukeModel](/docs/transformers/main/id/model_doc/luke#transformers.LukeModel) (model LukeConfig)
- **lw_detr** -- [LwDetrModel](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrModel) (model LwDetrConfig)
- **lxmert** -- [LxmertModel](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertModel) (model LxmertConfig)
- **m2m_100** -- [M2M100Model](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Model) (model M2M100Config)
- **mamba** -- [MambaModel](/docs/transformers/main/id/model_doc/mamba#transformers.MambaModel) (model MambaConfig)
- **mamba2** -- [Mamba2Model](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Model) (model Mamba2Config)
- **marian** -- [MarianModel](/docs/transformers/main/id/model_doc/marian#transformers.MarianModel) (model MarianConfig)
- **markuplm** -- [MarkupLMModel](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMModel) (model MarkupLMConfig)
- **mask2former** -- [Mask2FormerModel](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerModel) (model Mask2FormerConfig)
- **maskformer** -- [MaskFormerModel](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerModel) (model MaskFormerConfig)
- **maskformer-swin** -- `MaskFormerSwinModel` (model MaskFormerSwinConfig)
- **mbart** -- [MBartModel](/docs/transformers/main/id/model_doc/mbart#transformers.MBartModel) (model MBartConfig)
- **megatron-bert** -- [MegatronBertModel](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertModel) (model MegatronBertConfig)
- **mellum** -- [MellumModel](/docs/transformers/main/id/model_doc/mellum#transformers.MellumModel) (model MellumConfig)
- **metaclip_2** -- [MetaClip2Model](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Model) (model MetaClip2Config)
- **mgp-str** -- [MgpstrForSceneTextRecognition](/docs/transformers/main/id/model_doc/mgp-str#transformers.MgpstrForSceneTextRecognition) (model MgpstrConfig)
- **mimi** -- [MimiModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiModel) (model MimiConfig)
- **mimi_decoder** -- [MimiDecoderModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiDecoderModel) (model MimiDecoderConfig)
- **mimi_encoder** -- [MimiEncoderModel](/docs/transformers/main/id/model_doc/mimi#transformers.MimiEncoderModel) (model MimiEncoderConfig)
- **mimo_v2_flash** -- [MiMoV2FlashModel](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashModel) (model MiMoV2FlashConfig)
- **minicpm3** -- [MiniCPM3Model](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Model) (model MiniCPM3Config)
- **minicpm_ocr** -- [MiniCPMOcrModel](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrModel) (model MiniCPMOcrConfig)
- **minicpm_ocr_vision** -- [MiniCPMOcrVisionModel](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrVisionModel) (model MiniCPMOcrVisionConfig)
- **minicpmv4_6** -- [MiniCPMV4_6Model](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Model) (model MiniCPMV4_6Config)
- **minicpmv4_7** -- [MiniCPMV4_7Model](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Model) (model MiniCPMV4_7Config)
- **minicpmv4_7_vision** -- [MiniCPMV4_7VisionModel](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7VisionModel) (model MiniCPMV4_7VisionConfig)
- **minimax** -- [MiniMaxModel](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxModel) (model MiniMaxConfig)
- **minimax_m2** -- [MiniMaxM2Model](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2Model) (model MiniMaxM2Config)
- **minimax_m3_vl** -- [MiniMaxM3VLModel](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLModel) (model MiniMaxM3VLConfig)
- **minimax_m3_vl_text** -- [MiniMaxM3VLTextModel](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLTextModel) (model MiniMaxM3VLTextConfig)
- **ministral** -- [MinistralModel](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralModel) (model MinistralConfig)
- **ministral3** -- [Ministral3Model](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Model) (model Ministral3Config)
- **mistral** -- [MistralModel](/docs/transformers/main/id/model_doc/mistral#transformers.MistralModel) (model MistralConfig)
- **mistral3** -- [Mistral3Model](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Model) (model Mistral3Config)
- **mistral4** -- [Mistral4Model](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Model) (model Mistral4Config)
- **mistral_large4** -- [MistralLarge4Model](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Model) (model MistralLarge4Config)
- **mistral_large4_text** -- [MistralLarge4TextModel](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4TextModel) (model MistralLarge4TextConfig)
- **mixtral** -- [MixtralModel](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralModel) (model MixtralConfig)
- **mlcd** -- [MLCDVisionModel](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionModel) (model MLCDVisionConfig)
- **mlcd_vision_model** -- [MLCDVisionModel](/docs/transformers/main/id/model_doc/mlcd#transformers.MLCDVisionModel) (model MLCDVisionConfig)
- **mllama** -- [MllamaModel](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaModel) (model MllamaConfig)
- **mm-grounding-dino** -- [MMGroundingDinoModel](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoModel) (model MMGroundingDinoConfig)
- **mobilebert** -- [MobileBertModel](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertModel) (model MobileBertConfig)
- **mobilenet_v1** -- [MobileNetV1Model](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1Model) (model MobileNetV1Config)
- **mobilenet_v2** -- [MobileNetV2Model](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Model) (model MobileNetV2Config)
- **mobilevit** -- [MobileViTModel](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTModel) (model MobileViTConfig)
- **mobilevitv2** -- [MobileViTV2Model](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Model) (model MobileViTV2Config)
- **modernbert** -- [ModernBertModel](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertModel) (model ModernBertConfig)
- **modernbert-decoder** -- [ModernBertDecoderModel](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderModel) (model ModernBertDecoderConfig)
- **modernvbert** -- [ModernVBertModel](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertModel) (model ModernVBertConfig)
- **molmo2** -- [Molmo2Model](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Model) (model Molmo2Config)
- **molmo2_text** -- [Molmo2TextModel](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2TextModel) (model Molmo2TextConfig)
- **moonshine** -- [MoonshineModel](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineModel) (model MoonshineConfig)
- **moonshine_streaming** -- [MoonshineStreamingModel](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingModel) (model MoonshineStreamingConfig)
- **moshi** -- [MoshiModel](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiModel) (model MoshiConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeModel](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeModel) (model MossTranscribeDiarizeConfig)
- **moss_transcribe_diarize_audio** -- [MossTranscribeDiarizeAudioModel](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeAudioModel) (model MossTranscribeDiarizeAudioConfig)
- **mpnet** -- [MPNetModel](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetModel) (model MPNetConfig)
- **mpt** -- [MptModel](/docs/transformers/main/id/model_doc/mpt#transformers.MptModel) (model MptConfig)
- **mra** -- [MraModel](/docs/transformers/main/id/model_doc/mra#transformers.MraModel) (model MraConfig)
- **mt5** -- [MT5Model](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Model) (model MT5Config)
- **muse_glimmer** -- [MuseGlimmerModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerModel) (model MuseGlimmerConfig)
- **muse_glimmer_assistant** -- [MuseGlimmerAssistantModel](/docs/transformers/main/id/model_doc/muse_glimmer_assistant#transformers.MuseGlimmerAssistantModel) (model MuseGlimmerAssistantConfig)
- **muse_glimmer_text** -- [MuseGlimmerTextModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerTextModel) (model MuseGlimmerTextConfig)
- **muse_glimmer_vision** -- [MuseGlimmerVisionModel](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerVisionModel) (model MuseGlimmerVisionConfig)
- **muse_spark** -- [MuseSparkModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkModel) (model MuseSparkConfig)
- **muse_spark_audio** -- [MuseSparkAudioModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkAudioModel) (model MuseSparkAudioConfig)
- **muse_spark_text** -- [MuseSparkTextModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkTextModel) (model MuseSparkTextConfig)
- **muse_spark_vision** -- [MuseSparkVisionModel](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkVisionModel) (model MuseSparkVisionConfig)
- **musicflamingo** -- [MusicFlamingoModel](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoModel) (model MusicFlamingoConfig)
- **musicgen** -- [MusicgenModel](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenModel) (model MusicgenConfig)
- **musicgen_melody** -- [MusicgenMelodyModel](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyModel) (model MusicgenMelodyConfig)
- **mvp** -- [MvpModel](/docs/transformers/main/id/model_doc/mvp#transformers.MvpModel) (model MvpConfig)
- **nanochat** -- [NanoChatModel](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatModel) (model NanoChatConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeModel) (model Nemotron3_5TranscribeConfig)
- **nemotron** -- [NemotronModel](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronModel) (model NemotronConfig)
- **nemotron3_5_asr** -- [Nemotron3_5AsrForRNNT](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrForRNNT) (model Nemotron3_5AsrConfig)
- **nemotron3_5_transcribe_asr_audio** -- [Nemotron3_5TranscribeAsrAudioModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAsrAudioModel) (model Nemotron3_5TranscribeAsrAudioConfig)
- **nemotron3_5_transcribe_audio** -- [Nemotron3_5TranscribeAudioModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeAudioModel) (model Nemotron3_5TranscribeAudioConfig)
- **nemotron3_5_transcribe_word_timestamp** -- [Nemotron3_5TranscribeWordTimestampModel](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeWordTimestampModel) (model Nemotron3_5TranscribeWordTimestampConfig)
- **nemotron3_diarization** -- [Nemotron3DiarizationModel](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationModel) (model Nemotron3DiarizationConfig)
- **nemotron3_diarization_audio** -- [Nemotron3DiarizationAudioModel](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationAudioModel) (model Nemotron3DiarizationAudioConfig)
- **nemotron_asr_streaming** -- [NemotronAsrStreamingForRNNT](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingForRNNT) (model NemotronAsrStreamingConfig)
- **nemotron_asr_streaming_encoder** -- [NemotronAsrStreamingEncoder](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingEncoder) (model NemotronAsrStreamingEncoderConfig)
- **nemotron_h** -- [NemotronHModel](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHModel) (model NemotronHConfig)
- **neomme** -- [NeoMMEModel](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEModel) (model NeoMMEConfig)
- **neucodec** -- [NeuCodecModel](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecModel) (model NeuCodecConfig)
- **nllb-moe** -- [NllbMoeModel](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeModel) (model NllbMoeConfig)
- **nomic_bert** -- [NomicBertModel](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertModel) (model NomicBertConfig)
- **nystromformer** -- [NystromformerModel](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerModel) (model NystromformerConfig)
- **olmo** -- [OlmoModel](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoModel) (model OlmoConfig)
- **olmo2** -- [Olmo2Model](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Model) (model Olmo2Config)
- **olmo3** -- [Olmo3Model](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Model) (model Olmo3Config)
- **olmo_hybrid** -- [OlmoHybridModel](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridModel) (model OlmoHybridConfig)
- **olmoe** -- [OlmoeModel](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeModel) (model OlmoeConfig)
- **omdet-turbo** -- [OmDetTurboForObjectDetection](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboForObjectDetection) (model OmDetTurboConfig)
- **oneformer** -- [OneFormerModel](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerModel) (model OneFormerConfig)
- **openai-gpt** -- [OpenAIGPTModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTModel) (model OpenAIGPTConfig)
- **openai_privacy_filter** -- [OpenAIPrivacyFilterModel](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterModel) (model OpenAIPrivacyFilterConfig)
- **opt** -- [OPTModel](/docs/transformers/main/id/model_doc/opt#transformers.OPTModel) (model OPTConfig)
- **ovis2** -- [Ovis2Model](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Model) (model Ovis2Config)
- **owlv2** -- [Owlv2Model](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Model) (model Owlv2Config)
- **owlvit** -- [OwlViTModel](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTModel) (model OwlViTConfig)
- **paddleocr_vl** -- [PaddleOCRVLModel](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLModel) (model PaddleOCRVLConfig)
- **paligemma** -- [PaliGemmaModel](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaModel) (model PaliGemmaConfig)
- **parakeet_ctc** -- [ParakeetForCTC](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForCTC) (model ParakeetCTCConfig)
- **parakeet_encoder** -- [ParakeetEncoder](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetEncoder) (model ParakeetEncoderConfig)
- **parakeet_rnnt** -- [ParakeetForRNNT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForRNNT) (model ParakeetRNNTConfig)
- **parakeet_tdt** -- [ParakeetForTDT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForTDT) (model ParakeetTDTConfig)
- **patchtsmixer** -- [PatchTSMixerModel](/docs/transformers/main/id/model_doc/patchtsmixer#transformers.PatchTSMixerModel) (model PatchTSMixerConfig)
- **patchtst** -- [PatchTSTModel](/docs/transformers/main/id/model_doc/patchtst#transformers.PatchTSTModel) (model PatchTSTConfig)
- **pe_audio** -- [PeAudioModel](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioModel) (model PeAudioConfig)
- **pe_audio_encoder** -- [PeAudioEncoder](/docs/transformers/main/id/model_doc/pe_audio#transformers.PeAudioEncoder) (model PeAudioEncoderConfig)
- **pe_audio_video** -- [PeAudioVideoModel](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoModel) (model PeAudioVideoConfig)
- **pe_audio_video_encoder** -- [PeAudioVideoEncoder](/docs/transformers/main/id/model_doc/pe_audio_video#transformers.PeAudioVideoEncoder) (model PeAudioVideoEncoderConfig)
- **pe_video** -- [PeVideoModel](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoModel) (model PeVideoConfig)
- **pe_video_encoder** -- [PeVideoEncoder](/docs/transformers/main/id/model_doc/pe_video#transformers.PeVideoEncoder) (model PeVideoEncoderConfig)
- **pegasus** -- [PegasusModel](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusModel) (model PegasusConfig)
- **pegasus_x** -- [PegasusXModel](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXModel) (model PegasusXConfig)
- **perceiver** -- [PerceiverModel](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverModel) (model PerceiverConfig)
- **perception_lm** -- [PerceptionLMModel](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMModel) (model PerceptionLMConfig)
- **persimmon** -- [PersimmonModel](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonModel) (model PersimmonConfig)
- **phi** -- [PhiModel](/docs/transformers/main/id/model_doc/phi#transformers.PhiModel) (model PhiConfig)
- **phi3** -- [Phi3Model](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Model) (model Phi3Config)
- **phi4_multimodal** -- [Phi4MultimodalModel](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalModel) (model Phi4MultimodalConfig)
- **phimoe** -- [PhimoeModel](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeModel) (model PhimoeConfig)
- **pi0** -- [PI0Model](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Model) (model PI0Config)
- **pixio** -- [PixioModel](/docs/transformers/main/id/model_doc/pixio#transformers.PixioModel) (model PixioConfig)
- **pixtral** -- [PixtralVisionModel](/docs/transformers/main/id/model_doc/pixtral#transformers.PixtralVisionModel) (model PixtralVisionConfig)
- **plbart** -- [PLBartModel](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartModel) (model PLBartConfig)
- **pocket_tts** -- [PocketTTSModel](/docs/transformers/main/id/model_doc/pocket_tts#transformers.PocketTTSModel) (model PocketTTSConfig)
- **pocket_tts_audio_tokenizer** -- [PocketTTSAudioTokenizerModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerModel) (model PocketTTSAudioTokenizerConfig)
- **pocket_tts_audio_tokenizer_decoder** -- [PocketTTSAudioTokenizerDecoderModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerDecoderModel) (model PocketTTSAudioTokenizerDecoderConfig)
- **pocket_tts_audio_tokenizer_encoder** -- [PocketTTSAudioTokenizerEncoderModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerEncoderModel) (model PocketTTSAudioTokenizerEncoderConfig)
- **poolformer** -- [PoolFormerModel](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerModel) (model PoolFormerConfig)
- **pp_doclayout_v3** -- [PPDocLayoutV3Model](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3Model) (model PPDocLayoutV3Config)
- **pp_doclayout_v4** -- [PPDocLayoutV4Model](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4Model) (model PPDocLayoutV4Config)
- **pp_ocrv5_mobile_rec** -- [PPOCRV5MobileRecModel](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecModel) (model PPOCRV5MobileRecConfig)
- **pp_ocrv5_server_rec** -- [PPOCRV5ServerRecModel](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecModel) (model PPOCRV5ServerRecConfig)
- **pp_ocrv6_medium_det** -- [PPOCRV6MediumDetModel](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetModel) (model PPOCRV6MediumDetConfig)
- **pp_ocrv6_small_det** -- [PPOCRV6SmallDetModel](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetModel) (model PPOCRV6SmallDetConfig)
- **pp_ocrv6_small_rec** -- [PPOCRV6SmallRecModel](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecModel) (model PPOCRV6SmallRecConfig)
- **pp_ocrv6_tiny_rec** -- [PPOCRV6TinyRecModel](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecModel) (model PPOCRV6TinyRecConfig)
- **prophetnet** -- [ProphetNetModel](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetModel) (model ProphetNetConfig)
- **pvt** -- [PvtModel](/docs/transformers/main/id/model_doc/pvt#transformers.PvtModel) (model PvtConfig)
- **pvt_v2** -- [PvtV2Model](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2Model) (model PvtV2Config)
- **qianfan_ocr** -- [QianfanOCRModel](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRModel) (model QianfanOCRConfig)
- **qianfan_ocr_vision** -- [QianfanOCRVisionModel](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRVisionModel) (model QianfanOCRVisionConfig)
- **qwen2** -- [Qwen2Model](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Model) (model Qwen2Config)
- **qwen2_5_omni_thinker** -- [Qwen2_5OmniThinkerModel](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerModel) (model Qwen2_5OmniThinkerConfig)
- **qwen2_5_vl** -- [Qwen2_5_VLModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLModel) (model Qwen2_5_VLConfig)
- **qwen2_5_vl_text** -- [Qwen2_5_VLTextModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLTextModel) (model Qwen2_5_VLTextConfig)
- **qwen2_5_vl_vision** -- [Qwen2_5_VisionTransformerPretrainedModel](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VisionTransformerPretrainedModel) (model Qwen2_5_VLVisionConfig)
- **qwen2_audio** -- [Qwen2AudioModel](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioModel) (model Qwen2AudioConfig)
- **qwen2_audio_encoder** -- [Qwen2AudioEncoder](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioEncoder) (model Qwen2AudioEncoderConfig)
- **qwen2_moe** -- [Qwen2MoeModel](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeModel) (model Qwen2MoeConfig)
- **qwen2_vl** -- [Qwen2VLModel](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLModel) (model Qwen2VLConfig)
- **qwen2_vl_text** -- [Qwen2VLTextModel](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLTextModel) (model Qwen2VLTextConfig)
- **qwen3** -- [Qwen3Model](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Model) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5Model](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Model) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5MoeModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeModel) (model Qwen3_5MoeConfig)
- **qwen3_5_moe_text** -- [Qwen3_5MoeTextModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeTextModel) (model Qwen3_5MoeTextConfig)
- **qwen3_5_moe_vision** -- [Qwen3_5MoeVisionModel](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeVisionModel) (model Qwen3_5MoeVisionConfig)
- **qwen3_5_text** -- [Qwen3_5TextModel](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextModel) (model Qwen3_5TextConfig)
- **qwen3_5_vision** -- [Qwen3_5VisionModel](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5VisionModel) (model Qwen3_5VisionConfig)
- **qwen3_asr** -- [Qwen3ASRModel](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRModel) (model Qwen3ASRConfig)
- **qwen3_asr_encoder** -- [Qwen3ASREncoder](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASREncoder) (model Qwen3ASREncoderConfig)
- **qwen3_moe** -- [Qwen3MoeModel](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeModel) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextModel](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextModel) (model Qwen3NextConfig)
- **qwen3_omni_moe_thinker** -- [Qwen3OmniMoeThinkerModel](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerModel) (model Qwen3OmniMoeThinkerConfig)
- **qwen3_vl** -- [Qwen3VLModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLModel) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen3VLMoeModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeModel) (model Qwen3VLMoeConfig)
- **qwen3_vl_moe_text** -- [Qwen3VLMoeTextModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeTextModel) (model Qwen3VLMoeTextConfig)
- **qwen3_vl_moe_vision** -- [Qwen3VLMoeVisionModel](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeVisionModel) (model Qwen3VLMoeVisionConfig)
- **qwen3_vl_text** -- [Qwen3VLTextModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLTextModel) (model Qwen3VLTextConfig)
- **qwen3_vl_vision** -- [Qwen3VLVisionModel](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLVisionModel) (model Qwen3VLVisionConfig)
- **qwen4_exp** -- [Qwen4ExpModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpModel) (model Qwen4ExpConfig)
- **qwen4_exp_text** -- [Qwen4ExpTextModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpTextModel) (model Qwen4ExpTextConfig)
- **qwen4_exp_vision** -- [Qwen4ExpVisionModel](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpVisionModel) (model Qwen4ExpVisionConfig)
- **radio** -- [RadioModel](/docs/transformers/main/id/model_doc/radio#transformers.RadioModel) (model RadioConfig)
- **recurrent_gemma** -- [RecurrentGemmaModel](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaModel) (model RecurrentGemmaConfig)
- **reformer** -- [ReformerModel](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerModel) (model ReformerConfig)
- **regnet** -- [RegNetModel](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetModel) (model RegNetConfig)
- **rembert** -- [RemBertModel](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertModel) (model RemBertConfig)
- **resnet** -- [ResNetModel](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetModel) (model ResNetConfig)
- **rf_detr** -- [RfDetrModel](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrModel) (model RfDetrConfig)
- **roberta** -- [RobertaModel](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaModel) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormModel](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormModel) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertModel](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertModel) (model RoCBertConfig)
- **roformer** -- [RoFormerModel](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerModel) (model RoFormerConfig)
- **rt_detr** -- [RTDetrModel](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrModel) (model RTDetrConfig)
- **rt_detr_v2** -- [RTDetrV2Model](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2Model) (model RTDetrV2Config)
- **rwkv** -- [RwkvModel](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvModel) (model RwkvConfig)
- **sam** -- [SamModel](/docs/transformers/main/id/model_doc/sam#transformers.SamModel) (model SamConfig)
- **sam2** -- [Sam2Model](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2Model) (model Sam2Config)
- **sam2_hiera_det_model** -- [Sam2HieraDetModel](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2HieraDetModel) (model Sam2HieraDetConfig)
- **sam2_video** -- [Sam2VideoModel](/docs/transformers/main/id/model_doc/sam2_video#transformers.Sam2VideoModel) (model Sam2VideoConfig)
- **sam2_vision_model** -- [Sam2VisionModel](/docs/transformers/main/id/model_doc/sam2#transformers.Sam2VisionModel) (model Sam2VisionConfig)
- **sam3** -- [Sam3Model](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3Model) (model Sam3Config)
- **sam3_lite_text** -- [Sam3LiteTextModel](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextModel) (model Sam3LiteTextConfig)
- **sam3_lite_text_text_model** -- [Sam3LiteTextTextModel](/docs/transformers/main/id/model_doc/sam3_lite_text#transformers.Sam3LiteTextTextModel) (model Sam3LiteTextTextConfig)
- **sam3_tracker** -- [Sam3TrackerModel](/docs/transformers/main/id/model_doc/sam3_tracker#transformers.Sam3TrackerModel) (model Sam3TrackerConfig)
- **sam3_tracker_video** -- [Sam3TrackerVideoModel](/docs/transformers/main/id/model_doc/sam3_tracker_video#transformers.Sam3TrackerVideoModel) (model Sam3TrackerVideoConfig)
- **sam3_video** -- [Sam3VideoModel](/docs/transformers/main/id/model_doc/sam3_video#transformers.Sam3VideoModel) (model Sam3VideoConfig)
- **sam3_vision_model** -- [Sam3VisionModel](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3VisionModel) (model Sam3VisionConfig)
- **sam3_vit_model** -- [Sam3ViTModel](/docs/transformers/main/id/model_doc/sam3#transformers.Sam3ViTModel) (model Sam3ViTConfig)
- **sam_hq** -- [SamHQModel](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQModel) (model SamHQConfig)
- **sam_hq_vision_model** -- [SamHQVisionModel](/docs/transformers/main/id/model_doc/sam_hq#transformers.SamHQVisionModel) (model SamHQVisionConfig)
- **sam_vision_model** -- [SamVisionModel](/docs/transformers/main/id/model_doc/sam#transformers.SamVisionModel) (model SamVisionConfig)
- **sapiens2** -- [Sapiens2Model](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Model) (model Sapiens2Config)
- **seamless_m4t** -- [SeamlessM4TModel](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TModel) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4Tv2Model](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Model) (model SeamlessM4Tv2Config)
- **seed_oss** -- [SeedOssModel](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssModel) (model SeedOssConfig)
- **segformer** -- [SegformerModel](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerModel) (model SegformerConfig)
- **seggpt** -- [SegGptModel](/docs/transformers/main/id/model_doc/seggpt#transformers.SegGptModel) (model SegGptConfig)
- **sew** -- [SEWModel](/docs/transformers/main/id/model_doc/sew#transformers.SEWModel) (model SEWConfig)
- **sew-d** -- [SEWDModel](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDModel) (model SEWDConfig)
- **siglip** -- [SiglipModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipModel) (model SiglipConfig)
- **siglip2** -- [Siglip2Model](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Model) (model Siglip2Config)
- **siglip2_vision_model** -- [Siglip2VisionModel](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2VisionModel) (model Siglip2VisionConfig)
- **siglip_vision_model** -- [SiglipVisionModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipVisionModel) (model SiglipVisionConfig)
- **smollm3** -- [SmolLM3Model](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Model) (model SmolLM3Config)
- **smolvlm** -- [SmolVLMModel](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMModel) (model SmolVLMConfig)
- **smolvlm_vision** -- [SmolVLMVisionTransformer](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMVisionTransformer) (model SmolVLMVisionConfig)
- **solar_open** -- [SolarOpenModel](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenModel) (model SolarOpenConfig)
- **speech_to_text** -- [Speech2TextModel](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextModel) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5Model](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Model) (model SpeechT5Config)
- **splinter** -- [SplinterModel](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterModel) (model SplinterConfig)
- **squeezebert** -- [SqueezeBertModel](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertModel) (model SqueezeBertConfig)
- **stablelm** -- [StableLmModel](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmModel) (model StableLmConfig)
- **starcoder2** -- [Starcoder2Model](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Model) (model Starcoder2Config)
- **step3p7** -- [Step3p7Model](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Model) (model Step3p7Config)
- **swiftformer** -- [SwiftFormerModel](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerModel) (model SwiftFormerConfig)
- **swin** -- [SwinModel](/docs/transformers/main/id/model_doc/swin#transformers.SwinModel) (model SwinConfig)
- **swin2sr** -- [Swin2SRModel](/docs/transformers/main/id/model_doc/swin2sr#transformers.Swin2SRModel) (model Swin2SRConfig)
- **swinv2** -- [Swinv2Model](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Model) (model Swinv2Config)
- **switch_transformers** -- [SwitchTransformersModel](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersModel) (model SwitchTransformersConfig)
- **t5** -- [T5Model](/docs/transformers/main/id/model_doc/t5#transformers.T5Model) (model T5Config)
- **t5gemma** -- [T5GemmaModel](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaModel) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2Model](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Model) (model T5Gemma2Config)
- **t5gemma2_encoder** -- `T5Gemma2Encoder` (model T5Gemma2EncoderConfig)
- **table-transformer** -- [TableTransformerModel](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerModel) (model TableTransformerConfig)
- **tapas** -- [TapasModel](/docs/transformers/main/id/model_doc/tapas#transformers.TapasModel) (model TapasConfig)
- **textnet** -- [TextNetModel](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetModel) (model TextNetConfig)
- **time_series_transformer** -- [TimeSeriesTransformerModel](/docs/transformers/main/id/model_doc/time_series_transformer#transformers.TimeSeriesTransformerModel) (model TimeSeriesTransformerConfig)
- **timesfm** -- [TimesFmModel](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmModel) (model TimesFmConfig)
- **timesfm2_5** -- [TimesFm2_5Model](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5Model) (model TimesFm2_5Config)
- **timesformer** -- [TimesformerModel](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerModel) (model TimesformerConfig)
- **timm_backbone** -- [TimmBackbone](/docs/transformers/main/id/main_classes/backbones#transformers.TimmBackbone) (model TimmBackboneConfig)
- **timm_wrapper** -- [TimmWrapperModel](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperModel) (model TimmWrapperConfig)
- **tipsv2** -- [Tipsv2Model](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Model) (model Tipsv2Config)
- **tipsv2_text_model** -- [Tipsv2TextModel](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2TextModel) (model Tipsv2TextConfig)
- **tipsv2_vision_model** -- [Tipsv2VisionModel](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2VisionModel) (model Tipsv2VisionConfig)
- **tvp** -- [TvpModel](/docs/transformers/main/id/model_doc/tvp#transformers.TvpModel) (model TvpConfig)
- **udop** -- [UdopModel](/docs/transformers/main/id/model_doc/udop#transformers.UdopModel) (model UdopConfig)
- **ultrabert** -- [UltrabertModel](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertModel) (model UltrabertConfig)
- **ultravbert** -- [UltraVBertModel](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertModel) (model UltraVBertConfig)
- **umt5** -- [UMT5Model](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Model) (model UMT5Config)
- **unispeech** -- [UniSpeechModel](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechModel) (model UniSpeechConfig)
- **unispeech-sat** -- [UniSpeechSatModel](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatModel) (model UniSpeechSatConfig)
- **univnet** -- [UnivNetModel](/docs/transformers/main/id/model_doc/univnet#transformers.UnivNetModel) (model UnivNetConfig)
- **unlimited_ocr** -- [UnlimitedOcrModel](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrModel) (model UnlimitedOcrConfig)
- **uvdoc** -- [UVDocModel](/docs/transformers/main/id/model_doc/uvdoc#transformers.UVDocModel) (model UVDocConfig)
- **vaultgemma** -- [VaultGemmaModel](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaModel) (model VaultGemmaConfig)
- **vibevoice** -- [VibeVoiceForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceForConditionalGeneration) (model VibeVoiceConfig)
- **vibevoice_acoustic_tokenizer** -- [VibeVoiceAcousticTokenizerModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerModel) (model VibeVoiceAcousticTokenizerConfig)
- **vibevoice_acoustic_tokenizer_decoder** -- [VibeVoiceAcousticTokenizerDecoderModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerDecoderModel) (model VibeVoiceAcousticTokenizerDecoderConfig)
- **vibevoice_acoustic_tokenizer_encoder** -- [VibeVoiceAcousticTokenizerEncoderModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerEncoderModel) (model VibeVoiceAcousticTokenizerEncoderConfig)
- **vibevoice_asr** -- [VibeVoiceAsrModel](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrModel) (model VibeVoiceAsrConfig)
- **video_llama_3** -- [VideoLlama3Model](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Model) (model VideoLlama3Config)
- **video_llama_3_vision** -- [VideoLlama3VisionModel](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3VisionModel) (model VideoLlama3VisionConfig)
- **video_llava** -- [VideoLlavaModel](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaModel) (model VideoLlavaConfig)
- **videomae** -- [VideoMAEModel](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEModel) (model VideoMAEConfig)
- **videoprism** -- [VideoPrismClipModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismClipModel) (model VideoPrismConfig)
- **videoprism_vision_model** -- [VideoPrismVisionModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismVisionModel) (model VideoPrismVisionConfig)
- **vilt** -- [ViltModel](/docs/transformers/main/id/model_doc/vilt#transformers.ViltModel) (model ViltConfig)
- **vipllava** -- [VipLlavaModel](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaModel) (model VipLlavaConfig)
- **vision-text-dual-encoder** -- [VisionTextDualEncoderModel](/docs/transformers/main/id/model_doc/vision-text-dual-encoder#transformers.VisionTextDualEncoderModel) (model VisionTextDualEncoderConfig)
- **visual_bert** -- [VisualBertModel](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertModel) (model VisualBertConfig)
- **vit** -- [ViTModel](/docs/transformers/main/id/model_doc/vit#transformers.ViTModel) (model ViTConfig)
- **vit_mae** -- [ViTMAEModel](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEModel) (model ViTMAEConfig)
- **vit_msn** -- [ViTMSNModel](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNModel) (model ViTMSNConfig)
- **vitdet** -- [VitDetModel](/docs/transformers/main/id/model_doc/vitdet#transformers.VitDetModel) (model VitDetConfig)
- **vits** -- [VitsModel](/docs/transformers/main/id/model_doc/vits#transformers.VitsModel) (model VitsConfig)
- **vivit** -- [VivitModel](/docs/transformers/main/id/model_doc/vivit#transformers.VivitModel) (model VivitConfig)
- **vjepa2** -- [VJEPA2Model](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2Model) (model VJEPA2Config)
- **voxtral** -- [VoxtralModel](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralModel) (model VoxtralConfig)
- **voxtral_encoder** -- [VoxtralEncoder](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralEncoder) (model VoxtralEncoderConfig)
- **voxtral_realtime** -- [VoxtralRealtimeModel](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeModel) (model VoxtralRealtimeConfig)
- **voxtral_realtime_encoder** -- [VoxtralRealtimeEncoder](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeEncoder) (model VoxtralRealtimeEncoderConfig)
- **voxtral_realtime_text** -- `VoxtralRealtimeTextModel` (model VoxtralRealtimeTextConfig)
- **wav2vec2** -- [Wav2Vec2Model](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Model) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertModel](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertModel) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerModel](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerModel) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMModel](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMModel) (model WavLMConfig)
- **whisper** -- [WhisperModel](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperModel) (model WhisperConfig)
- **xclip** -- [XCLIPModel](/docs/transformers/main/id/model_doc/xclip#transformers.XCLIPModel) (model XCLIPConfig)
- **xcodec** -- [XcodecModel](/docs/transformers/main/id/model_doc/xcodec#transformers.XcodecModel) (model XcodecConfig)
- **xcodec2** -- [Xcodec2Model](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Model) (model Xcodec2Config)
- **xglm** -- [XGLMModel](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMModel) (model XGLMConfig)
- **xlm** -- [XLMModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMModel) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaModel](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaModel) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLModel](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLModel) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetModel) (model XLNetConfig)
- **xlstm** -- [xLSTMModel](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMModel) (model xLSTMConfig)
- **xmod** -- [XmodModel](/docs/transformers/main/id/model_doc/xmod#transformers.XmodModel) (model XmodConfig)
- **yolos** -- [YolosModel](/docs/transformers/main/id/model_doc/yolos#transformers.YolosModel) (model YolosConfig)
- **yoso** -- [YosoModel](/docs/transformers/main/id/model_doc/yoso#transformers.YosoModel) (model YosoConfig)
- **youtu** -- [YoutuModel](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuModel) (model YoutuConfig)
- **zamba** -- [ZambaModel](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaModel) (model ZambaConfig)
- **zamba2** -- [Zamba2Model](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Model) (model Zamba2Config)
- **zaya** -- [ZayaModel](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaModel) (model ZayaConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModel

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModel.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModel.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

## Kelas pra-pelatihan generik

Kelas auto berikut tersedia untuk menginstansiasi model dengan kepala pra-pelatihan.

### AutoModelForPreTraining[[transformers.AutoModelForPreTraining]]

#### transformers.AutoModelForPreTraining[[transformers.AutoModelForPreTraining]]

```python
transformers.AutoModelForPreTraining(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2304)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pra-pelatihan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForPreTraining.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig) configuration class: [AlbertForPreTraining](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForPreTraining) (AlbertConfig model) - [AudioFlamingo3Config](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Config) configuration class: [AudioFlamingo3ForConditionalGeneration](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3ForConditionalGeneration) (AudioFlamingo3Config model) - [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig) configuration class: [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (BartConfig model) - [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) configuration class: [BertForPreTraining](/docs/transformers/main/id/model_doc/bert#transformers.BertForPreTraining) (BertConfig model) - [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) configuration class: [BigBirdForPreTraining](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForPreTraining) (BigBirdConfig model) - [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) configuration class: [BloomForCausalLM](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForCausalLM) (BloomConfig model) - [CTRLConfig](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLConfig) configuration class: [CTRLLMHeadModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLLMHeadModel) (CTRLConfig model) - [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) configuration class: [CamembertForMaskedLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMaskedLM) (CamembertConfig model) - [ColModernVBertConfig](/docs/transformers/main/id/model_doc/colmodernvbert#transformers.ColModernVBertConfig) configuration class: [ColModernVBertForRetrieval](/docs/transformers/main/id/model_doc/colmodernvbert#transformers.ColModernVBertForRetrieval) (ColModernVBertConfig model) - [ColPaliConfig](/docs/transformers/main/id/model_doc/colpali#transformers.ColPaliConfig) configuration class: [ColPaliForRetrieval](/docs/transformers/main/id/model_doc/colpali#transformers.ColPaliForRetrieval) (ColPaliConfig model) - [ColQwen2Config](/docs/transformers/main/id/model_doc/colqwen2#transformers.ColQwen2Config) configuration class: [ColQwen2ForRetrieval](/docs/transformers/main/id/model_doc/colqwen2#transformers.ColQwen2ForRetrieval) (ColQwen2Config model) - [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) configuration class: [Data2VecTextForMaskedLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMaskedLM) (Data2VecTextConfig model) - [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig) configuration class: [DebertaForMaskedLM](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForMaskedLM) (DebertaConfig model) - [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config) configuration class: [DebertaV2ForMaskedLM](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMaskedLM) (DebertaV2Config model) - [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig) configuration class: [DistilBertForMaskedLM](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMaskedLM) (DistilBertConfig model) - [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) configuration class: [ElectraForPreTraining](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForPreTraining) (ElectraConfig model) - [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) configuration class: [ErnieForPreTraining](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForPreTraining) (ErnieConfig model) - [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) configuration class: [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (EvollaConfig model) - [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) configuration class: [Exaone4ForCausalLM](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForCausalLM) (Exaone4Config model) - [ExaoneMoeConfig](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeConfig) configuration class: [ExaoneMoeForCausalLM](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeForCausalLM) (ExaoneMoeConfig model) - [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig) configuration class: [FNetForPreTraining](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForPreTraining) (FNetConfig model) - [FSMTConfig](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTConfig) configuration class: [FSMTForConditionalGeneration](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTForConditionalGeneration) (FSMTConfig model) - [FalconMambaConfig](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaConfig) configuration class: [FalconMambaForCausalLM](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaForCausalLM) (FalconMambaConfig model) - [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig) configuration class: [FlaubertWithLMHeadModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertWithLMHeadModel) (FlaubertConfig model) - [FlavaConfig](/docs/transformers/main/id/model_doc/flava#transformers.FlavaConfig) configuration class: [FlavaForPreTraining](/docs/transformers/main/id/model_doc/flava#transformers.FlavaForPreTraining) (FlavaConfig model) - [Florence2Config](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Config) configuration class: [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (Florence2Config model) - [FunAsrNanoConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoConfig) configuration class: [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (FunAsrNanoConfig model) - [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig) configuration class: [FunnelForPreTraining](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForPreTraining) (FunnelConfig model) - [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) configuration class: [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (GPT2Config model) - [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) configuration class: [GPTBigCodeForCausalLM](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForCausalLM) (GPTBigCodeConfig model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (Gemma3Config model) - [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) configuration class: [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (Gemma4Config model) - [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) configuration class: [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (Gemma4UnifiedConfig model) - [GlmAsrConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrConfig) configuration class: [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (GlmAsrConfig model) - [HieraConfig](/docs/transformers/main/id/model_doc/hiera#transformers.HieraConfig) configuration class: [HieraForPreTraining](/docs/transformers/main/id/model_doc/hiera#transformers.HieraForPreTraining) (HieraConfig model) - [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig) configuration class: [IBertForMaskedLM](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMaskedLM) (IBertConfig model) - [Idefics2Config](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Config) configuration class: [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (Idefics2Config model) - [Idefics3Config](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Config) configuration class: [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (Idefics3Config model) - [IdeficsConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsConfig) configuration class: [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (IdeficsConfig model) - [InklingConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingConfig) configuration class: [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (InklingConfig model) - [JanusConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusConfig) configuration class: [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (JanusConfig model) - [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig) configuration class: [LayoutLMForMaskedLM](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForMaskedLM) (LayoutLMConfig model) - [LlavaConfig](/docs/transformers/main/id/model_doc/llava#transformers.LlavaConfig) configuration class: [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (LlavaConfig model) - [LlavaNextConfig](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextConfig) configuration class: [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (LlavaNextConfig model) - [LlavaNextVideoConfig](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoConfig) configuration class: [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (LlavaNextVideoConfig model) - [LlavaOnevisionConfig](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionConfig) configuration class: [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (LlavaOnevisionConfig model) - [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig) configuration class: [LongformerForMaskedLM](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMaskedLM) (LongformerConfig model) - [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig) configuration class: [LukeForMaskedLM](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMaskedLM) (LukeConfig model) - [LxmertConfig](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertConfig) configuration class: [LxmertForPreTraining](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertForPreTraining) (LxmertConfig model) - [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig) configuration class: [MPNetForMaskedLM](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMaskedLM) (MPNetConfig model) - [Mamba2Config](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Config) configuration class: [Mamba2ForCausalLM](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2ForCausalLM) (Mamba2Config model) - [MambaConfig](/docs/transformers/main/id/model_doc/mamba#transformers.MambaConfig) configuration class: [MambaForCausalLM](/docs/transformers/main/id/model_doc/mamba#transformers.MambaForCausalLM) (MambaConfig model) - [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) configuration class: [MegatronBertForPreTraining](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForPreTraining) (MegatronBertConfig model) - [Mistral3Config](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Config) configuration class: [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (Mistral3Config model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (Mistral4Config model) - [MistralLarge4Config](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Config) configuration class: [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (MistralLarge4Config model) - [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) configuration class: [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (MllamaConfig model) - [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig) configuration class: [MobileBertForPreTraining](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForPreTraining) (MobileBertConfig model) - [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig) configuration class: [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (MossTranscribeDiarizeConfig model) - [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) configuration class: [MptForCausalLM](/docs/transformers/main/id/model_doc/mpt#transformers.MptForCausalLM) (MptConfig model) - [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig) configuration class: [MraForMaskedLM](/docs/transformers/main/id/model_doc/mra#transformers.MraForMaskedLM) (MraConfig model) - [MusicFlamingoConfig](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoConfig) configuration class: [MusicFlamingoForConditionalGeneration](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoForConditionalGeneration) (MusicFlamingoConfig model) - [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig) configuration class: [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (MvpConfig model) - [NanoChatConfig](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatConfig) configuration class: [NanoChatForCausalLM](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatForCausalLM) (NanoChatConfig model) - [Nemotron3_5TranscribeConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeConfig) configuration class: [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (Nemotron3_5TranscribeConfig model) - [NllbMoeConfig](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeConfig) configuration class: [NllbMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeForConditionalGeneration) (NllbMoeConfig model) - [OpenAIGPTConfig](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTConfig) configuration class: [OpenAIGPTLMHeadModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTLMHeadModel) (OpenAIGPTConfig model) - [PaliGemmaConfig](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaConfig) configuration class: [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (PaliGemmaConfig model) - [Qwen2AudioConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioConfig) configuration class: [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (Qwen2AudioConfig model) - [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig) configuration class: [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (Qwen3ASRConfig model) - [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) configuration class: [RoCBertForPreTraining](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForPreTraining) (RoCBertConfig model) - [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) configuration class: [RobertaForMaskedLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMaskedLM) (RobertaConfig model) - [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) configuration class: [RobertaPreLayerNormForMaskedLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMaskedLM) (RobertaPreLayerNormConfig model) - [RwkvConfig](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvConfig) configuration class: [RwkvForCausalLM](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvForCausalLM) (RwkvConfig model) - [SplinterConfig](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterConfig) configuration class: [SplinterForPreTraining](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterForPreTraining) (SplinterConfig model) - [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig) configuration class: [SqueezeBertForMaskedLM](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMaskedLM) (SqueezeBertConfig model) - [SwitchTransformersConfig](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersConfig) configuration class: [SwitchTransformersForConditionalGeneration](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersForConditionalGeneration) (SwitchTransformersConfig model) - [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config) configuration class: [T5ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5#transformers.T5ForConditionalGeneration) (T5Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (T5Gemma2Config model) - [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig) configuration class: [T5GemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForConditionalGeneration) (T5GemmaConfig model) - [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig) configuration class: [TapasForMaskedLM](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForMaskedLM) (TapasConfig model) - [UniSpeechConfig](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechConfig) configuration class: [UniSpeechForPreTraining](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForPreTraining) (UniSpeechConfig model) - [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig) configuration class: [UniSpeechSatForPreTraining](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForPreTraining) (UniSpeechSatConfig model) - [ViTMAEConfig](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEConfig) configuration class: [ViTMAEForPreTraining](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEForPreTraining) (ViTMAEConfig model) - [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig) configuration class: [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (VibeVoiceAsrConfig model) - [VideoLlavaConfig](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaConfig) configuration class: [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (VideoLlavaConfig model) - [VideoMAEConfig](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEConfig) configuration class: [VideoMAEForPreTraining](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEForPreTraining) (VideoMAEConfig model) - [VipLlavaConfig](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaConfig) configuration class: [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (VipLlavaConfig model) - [VisualBertConfig](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertConfig) configuration class: [VisualBertForPreTraining](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertForPreTraining) (VisualBertConfig model) - [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig) configuration class: [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (VoxtralConfig model) - [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig) configuration class: [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (VoxtralRealtimeConfig model) - [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config) configuration class: [Wav2Vec2ForPreTraining](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForPreTraining) (Wav2Vec2Config model) - [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig) configuration class: [Wav2Vec2ConformerForPreTraining](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForPreTraining) (Wav2Vec2ConformerConfig model) - [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) configuration class: [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (XLMConfig model) - [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) configuration class: [XLMRobertaForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMaskedLM) (XLMRobertaConfig model) - [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) configuration class: [XLMRobertaXLForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMaskedLM) (XLMRobertaXLConfig model) - [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) configuration class: [XLNetLMHeadModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetLMHeadModel) (XLNetConfig model) - [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) configuration class: [XmodForMaskedLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMaskedLM) (XmodConfig model) - [xLSTMConfig](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMConfig) configuration class: [xLSTMForCausalLM](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMForCausalLM) (xLSTMConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pra-pelatihan) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPreTraining

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForPreTraining.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForPreTraining.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pra-pelatihan) dari model pra-terlatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForPreTraining](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForPreTraining) (model AlbertConfig)
- **audioflamingo3** -- [AudioFlamingo3ForConditionalGeneration](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3ForConditionalGeneration) (model AudioFlamingo3Config)
- **bart** -- [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (model BartConfig)
- **bert** -- [BertForPreTraining](/docs/transformers/main/id/model_doc/bert#transformers.BertForPreTraining) (model BertConfig)
- **big_bird** -- [BigBirdForPreTraining](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForPreTraining) (model BigBirdConfig)
- **bloom** -- [BloomForCausalLM](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForCausalLM) (model BloomConfig)
- **camembert** -- [CamembertForMaskedLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMaskedLM) (model CamembertConfig)
- **colmodernvbert** -- [ColModernVBertForRetrieval](/docs/transformers/main/id/model_doc/colmodernvbert#transformers.ColModernVBertForRetrieval) (model ColModernVBertConfig)
- **colpali** -- [ColPaliForRetrieval](/docs/transformers/main/id/model_doc/colpali#transformers.ColPaliForRetrieval) (model ColPaliConfig)
- **colqwen2** -- [ColQwen2ForRetrieval](/docs/transformers/main/id/model_doc/colqwen2#transformers.ColQwen2ForRetrieval) (model ColQwen2Config)
- **ctrl** -- [CTRLLMHeadModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLLMHeadModel) (model CTRLConfig)
- **data2vec-text** -- [Data2VecTextForMaskedLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMaskedLM) (model Data2VecTextConfig)
- **deberta** -- [DebertaForMaskedLM](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForMaskedLM) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2ForMaskedLM](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMaskedLM) (model DebertaV2Config)
- **distilbert** -- [DistilBertForMaskedLM](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMaskedLM) (model DistilBertConfig)
- **electra** -- [ElectraForPreTraining](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForPreTraining) (model ElectraConfig)
- **ernie** -- [ErnieForPreTraining](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForPreTraining) (model ErnieConfig)
- **evolla** -- [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (model EvollaConfig)
- **exaone4** -- [Exaone4ForCausalLM](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForCausalLM) (model Exaone4Config)
- **exaone_moe** -- [ExaoneMoeForCausalLM](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeForCausalLM) (model ExaoneMoeConfig)
- **falcon_mamba** -- [FalconMambaForCausalLM](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaForCausalLM) (model FalconMambaConfig)
- **flaubert** -- [FlaubertWithLMHeadModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertWithLMHeadModel) (model FlaubertConfig)
- **flava** -- [FlavaForPreTraining](/docs/transformers/main/id/model_doc/flava#transformers.FlavaForPreTraining) (model FlavaConfig)
- **florence2** -- [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (model Florence2Config)
- **fnet** -- [FNetForPreTraining](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForPreTraining) (model FNetConfig)
- **fsmt** -- [FSMTForConditionalGeneration](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTForConditionalGeneration) (model FSMTConfig)
- **fun_asr_nano** -- [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (model FunAsrNanoConfig)
- **funnel** -- [FunnelForPreTraining](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForPreTraining) (model FunnelConfig)
- **gemma3** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model Gemma3Config)
- **gemma4** -- [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (model Gemma4Config)
- **gemma4_unified** -- [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (model Gemma4UnifiedConfig)
- **glmasr** -- [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (model GlmAsrConfig)
- **gpt-sw3** -- [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (model GPT2Config)
- **gpt2** -- [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeForCausalLM](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForCausalLM) (model GPTBigCodeConfig)
- **hiera** -- [HieraForPreTraining](/docs/transformers/main/id/model_doc/hiera#transformers.HieraForPreTraining) (model HieraConfig)
- **ibert** -- [IBertForMaskedLM](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMaskedLM) (model IBertConfig)
- **idefics** -- [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (model IdeficsConfig)
- **idefics2** -- [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (model Idefics2Config)
- **idefics3** -- [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (model Idefics3Config)
- **inkling_mm_model** -- [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (model InklingConfig)
- **janus** -- [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (model JanusConfig)
- **layoutlm** -- [LayoutLMForMaskedLM](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForMaskedLM) (model LayoutLMConfig)
- **llava** -- [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (model LlavaConfig)
- **llava_next** -- [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (model LlavaOnevisionConfig)
- **longformer** -- [LongformerForMaskedLM](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMaskedLM) (model LongformerConfig)
- **luke** -- [LukeForMaskedLM](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMaskedLM) (model LukeConfig)
- **lxmert** -- [LxmertForPreTraining](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertForPreTraining) (model LxmertConfig)
- **mamba** -- [MambaForCausalLM](/docs/transformers/main/id/model_doc/mamba#transformers.MambaForCausalLM) (model MambaConfig)
- **mamba2** -- [Mamba2ForCausalLM](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2ForCausalLM) (model Mamba2Config)
- **megatron-bert** -- [MegatronBertForPreTraining](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForPreTraining) (model MegatronBertConfig)
- **mistral3** -- [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (model Mistral3Config)
- **mistral4** -- [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (model Mistral4Config)
- **mistral_large4** -- [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (model MistralLarge4Config)
- **mllama** -- [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (model MllamaConfig)
- **mobilebert** -- [MobileBertForPreTraining](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForPreTraining) (model MobileBertConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (model MossTranscribeDiarizeConfig)
- **mpnet** -- [MPNetForMaskedLM](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMaskedLM) (model MPNetConfig)
- **mpt** -- [MptForCausalLM](/docs/transformers/main/id/model_doc/mpt#transformers.MptForCausalLM) (model MptConfig)
- **mra** -- [MraForMaskedLM](/docs/transformers/main/id/model_doc/mra#transformers.MraForMaskedLM) (model MraConfig)
- **musicflamingo** -- [MusicFlamingoForConditionalGeneration](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoForConditionalGeneration) (model MusicFlamingoConfig)
- **mvp** -- [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (model MvpConfig)
- **nanochat** -- [NanoChatForCausalLM](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatForCausalLM) (model NanoChatConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (model Nemotron3_5TranscribeConfig)
- **nllb-moe** -- [NllbMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeForConditionalGeneration) (model NllbMoeConfig)
- **openai-gpt** -- [OpenAIGPTLMHeadModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTLMHeadModel) (model OpenAIGPTConfig)
- **paligemma** -- [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (model PaliGemmaConfig)
- **qwen2_audio** -- [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (model Qwen2AudioConfig)
- **qwen3_asr** -- [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig)
- **roberta** -- [RobertaForMaskedLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMaskedLM) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForMaskedLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMaskedLM) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForPreTraining](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForPreTraining) (model RoCBertConfig)
- **rwkv** -- [RwkvForCausalLM](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvForCausalLM) (model RwkvConfig)
- **splinter** -- [SplinterForPreTraining](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterForPreTraining) (model SplinterConfig)
- **squeezebert** -- [SqueezeBertForMaskedLM](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMaskedLM) (model SqueezeBertConfig)
- **switch_transformers** -- [SwitchTransformersForConditionalGeneration](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersForConditionalGeneration) (model SwitchTransformersConfig)
- **t5** -- [T5ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5#transformers.T5ForConditionalGeneration) (model T5Config)
- **t5gemma** -- [T5GemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForConditionalGeneration) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (model T5Gemma2Config)
- **tapas** -- [TapasForMaskedLM](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForMaskedLM) (model TapasConfig)
- **unispeech** -- [UniSpeechForPreTraining](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForPreTraining) (model UniSpeechConfig)
- **unispeech-sat** -- [UniSpeechSatForPreTraining](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForPreTraining) (model UniSpeechSatConfig)
- **vibevoice_asr** -- [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig)
- **video_llava** -- [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (model VideoLlavaConfig)
- **videomae** -- [VideoMAEForPreTraining](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEForPreTraining) (model VideoMAEConfig)
- **vipllava** -- [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (model VipLlavaConfig)
- **visual_bert** -- [VisualBertForPreTraining](/docs/transformers/main/id/model_doc/visual_bert#transformers.VisualBertForPreTraining) (model VisualBertConfig)
- **vit_mae** -- [ViTMAEForPreTraining](/docs/transformers/main/id/model_doc/vit_mae#transformers.ViTMAEForPreTraining) (model ViTMAEConfig)
- **voxtral** -- [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig)
- **wav2vec2** -- [Wav2Vec2ForPreTraining](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForPreTraining) (model Wav2Vec2Config)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerForPreTraining](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForPreTraining) (model Wav2Vec2ConformerConfig)
- **xlm** -- [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMaskedLM) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMaskedLM) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetLMHeadModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetLMHeadModel) (model XLNetConfig)
- **xlstm** -- [xLSTMForCausalLM](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMForCausalLM) (model xLSTMConfig)
- **xmod** -- [XmodForMaskedLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMaskedLM) (model XmodConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPreTraining

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForPreTraining.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForPreTraining.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

## Pemrosesan Bahasa Alami

Kelas auto berikut tersedia untuk tugas pemrosesan bahasa alami berikut.

### AutoModelForCausalLM[[transformers.AutoModelForCausalLM]]

#### transformers.AutoModelForCausalLM[[transformers.AutoModelForCausalLM]]

```python
transformers.AutoModelForCausalLM(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2311)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa kausal) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForCausalLM.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AXK1Config](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Config) configuration class: [AXK1ForCausalLM](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForCausalLM) (AXK1Config model) - [AXK2Config](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Config) configuration class: [AXK2ForCausalLM](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForCausalLM) (AXK2Config model) - [AfmoeConfig](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeConfig) configuration class: [AfmoeForCausalLM](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeForCausalLM) (AfmoeConfig model) - [ApertusConfig](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusConfig) configuration class: [ApertusForCausalLM](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusForCausalLM) (ApertusConfig model) - [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig) configuration class: [ArceeForCausalLM](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForCausalLM) (ArceeConfig model) - [AriaTextConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextConfig) configuration class: [AriaTextForCausalLM](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextForCausalLM) (AriaTextConfig model) - [BailingHybridConfig](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridConfig) configuration class: [BailingHybridForCausalLM](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridForCausalLM) (BailingHybridConfig model) - [BambaConfig](/docs/transformers/main/id/model_doc/bamba#transformers.BambaConfig) configuration class: [BambaForCausalLM](/docs/transformers/main/id/model_doc/bamba#transformers.BambaForCausalLM) (BambaConfig model) - [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig) configuration class: [BartForCausalLM](/docs/transformers/main/id/model_doc/bart#transformers.BartForCausalLM) (BartConfig model) - [BeamConfig](/docs/transformers/main/id/model_doc/beam#transformers.BeamConfig) configuration class: [BeamForCausalLM](/docs/transformers/main/id/model_doc/beam#transformers.BeamForCausalLM) (BeamConfig model) - [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) configuration class: [BertLMHeadModel](/docs/transformers/main/id/model_doc/bert#transformers.BertLMHeadModel) (BertConfig model) - [BertGenerationConfig](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationConfig) configuration class: [BertGenerationDecoder](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationDecoder) (BertGenerationConfig model) - [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) configuration class: [BigBirdForCausalLM](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForCausalLM) (BigBirdConfig model) - [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig) configuration class: [BigBirdPegasusForCausalLM](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForCausalLM) (BigBirdPegasusConfig model) - [BioGptConfig](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptConfig) configuration class: [BioGptForCausalLM](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForCausalLM) (BioGptConfig model) - [BitNetConfig](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetConfig) configuration class: [BitNetForCausalLM](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetForCausalLM) (BitNetConfig model) - [BlenderbotConfig](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotConfig) configuration class: [BlenderbotForCausalLM](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotForCausalLM) (BlenderbotConfig model) - [BlenderbotSmallConfig](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallConfig) configuration class: [BlenderbotSmallForCausalLM](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallForCausalLM) (BlenderbotSmallConfig model) - [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) configuration class: [BloomForCausalLM](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForCausalLM) (BloomConfig model) - [BltConfig](/docs/transformers/main/id/model_doc/blt#transformers.BltConfig) configuration class: [BltForCausalLM](/docs/transformers/main/id/model_doc/blt#transformers.BltForCausalLM) (BltConfig model) - [CTRLConfig](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLConfig) configuration class: [CTRLLMHeadModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLLMHeadModel) (CTRLConfig model) - [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) configuration class: [CamembertForCausalLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForCausalLM) (CamembertConfig model) - [CodeGenConfig](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenConfig) configuration class: [CodeGenForCausalLM](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenForCausalLM) (CodeGenConfig model) - [Cohere2Config](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2Config) configuration class: [Cohere2ForCausalLM](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2ForCausalLM) (Cohere2Config model) - [Cohere2MoeConfig](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeConfig) configuration class: [Cohere2MoeForCausalLM](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeForCausalLM) (Cohere2MoeConfig model) - [CohereCompassTextConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextConfig) configuration class: [CohereCompassForCausalLM](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForCausalLM) (CohereCompassTextConfig model) - [CohereConfig](/docs/transformers/main/id/model_doc/cohere#transformers.CohereConfig) configuration class: [CohereForCausalLM](/docs/transformers/main/id/model_doc/cohere#transformers.CohereForCausalLM) (CohereConfig model) - [CpmAntConfig](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntConfig) configuration class: [CpmAntForCausalLM](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntForCausalLM) (CpmAntConfig model) - [CwmConfig](/docs/transformers/main/id/model_doc/cwm#transformers.CwmConfig) configuration class: [CwmForCausalLM](/docs/transformers/main/id/model_doc/cwm#transformers.CwmForCausalLM) (CwmConfig model) - [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) configuration class: [Data2VecTextForCausalLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForCausalLM) (Data2VecTextConfig model) - [DbrxConfig](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxConfig) configuration class: [DbrxForCausalLM](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxForCausalLM) (DbrxConfig model) - [DeepseekV2Config](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Config) configuration class: [DeepseekV2ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2ForCausalLM) (DeepseekV2Config model) - [DeepseekV32Config](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32Config) configuration class: [DeepseekV32ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32ForCausalLM) (DeepseekV32Config model) - [DeepseekV3Config](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Config) configuration class: [DeepseekV3ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForCausalLM) (DeepseekV3Config model) - [DeepseekV4Config](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4Config) configuration class: [DeepseekV4ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4ForCausalLM) (DeepseekV4Config model) - [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig) configuration class: [DiffLlamaForCausalLM](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForCausalLM) (DiffLlamaConfig model) - [DogeConfig](/docs/transformers/main/id/model_doc/doge#transformers.DogeConfig) configuration class: [DogeForCausalLM](/docs/transformers/main/id/model_doc/doge#transformers.DogeForCausalLM) (DogeConfig model) - [Dots1Config](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1Config) configuration class: [Dots1ForCausalLM](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1ForCausalLM) (Dots1Config model) - [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) configuration class: [ElectraForCausalLM](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForCausalLM) (ElectraConfig model) - [Emu3Config](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Config) configuration class: [Emu3ForCausalLM](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForCausalLM) (Emu3Config model) - [Ernie4_5Config](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5Config) configuration class: [Ernie4_5ForCausalLM](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5ForCausalLM) (Ernie4_5Config model) - [Ernie4_5_MoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeConfig) configuration class: [Ernie4_5_MoeForCausalLM](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeForCausalLM) (Ernie4_5_MoeConfig model) - [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) configuration class: [ErnieForCausalLM](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForCausalLM) (ErnieConfig model) - [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) configuration class: [Exaone4ForCausalLM](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForCausalLM) (Exaone4Config model) - [ExaoneMoeConfig](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeConfig) configuration class: [ExaoneMoeForCausalLM](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeForCausalLM) (ExaoneMoeConfig model) - [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig) configuration class: [FalconForCausalLM](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForCausalLM) (FalconConfig model) - [FalconH1Config](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1Config) configuration class: [FalconH1ForCausalLM](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1ForCausalLM) (FalconH1Config model) - [FalconMambaConfig](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaConfig) configuration class: [FalconMambaForCausalLM](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaForCausalLM) (FalconMambaConfig model) - [FlexOlmoConfig](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoConfig) configuration class: [FlexOlmoForCausalLM](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoForCausalLM) (FlexOlmoConfig model) - [FlintConfig](/docs/transformers/main/id/model_doc/flint#transformers.FlintConfig) configuration class: [FlintForCausalLM](/docs/transformers/main/id/model_doc/flint#transformers.FlintForCausalLM) (FlintConfig model) - [FuyuConfig](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuConfig) configuration class: [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (FuyuConfig model) - [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) configuration class: [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (GPT2Config model) - [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) configuration class: [GPTBigCodeForCausalLM](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForCausalLM) (GPTBigCodeConfig model) - [GPTJConfig](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJConfig) configuration class: [GPTJForCausalLM](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForCausalLM) (GPTJConfig model) - [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig) configuration class: [GPTNeoForCausalLM](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForCausalLM) (GPTNeoConfig model) - [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig) configuration class: [GPTNeoXForCausalLM](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForCausalLM) (GPTNeoXConfig model) - [GPTNeoXJapaneseConfig](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseConfig) configuration class: [GPTNeoXJapaneseForCausalLM](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseForCausalLM) (GPTNeoXJapaneseConfig model) - [Gemma2Config](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Config) configuration class: [Gemma2ForCausalLM](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForCausalLM) (Gemma2Config model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (Gemma3Config model) - [Gemma3TextConfig](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextConfig) configuration class: [Gemma3ForCausalLM](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForCausalLM) (Gemma3TextConfig model) - [Gemma3nConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nConfig) configuration class: [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (Gemma3nConfig model) - [Gemma3nTextConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nTextConfig) configuration class: [Gemma3nForCausalLM](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForCausalLM) (Gemma3nTextConfig model) - [Gemma4AssistantConfig](/docs/transformers/main/id/model_doc/gemma4_assistant#transformers.Gemma4AssistantConfig) configuration class: [Gemma4AssistantForCausalLM](/docs/transformers/main/id/model_doc/gemma4_assistant#transformers.Gemma4AssistantForCausalLM) (Gemma4AssistantConfig model) - [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) configuration class: [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (Gemma4Config model) - [Gemma4TextConfig](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4TextConfig) configuration class: [Gemma4ForCausalLM](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForCausalLM) (Gemma4TextConfig model) - [Gemma4UnifiedAssistantConfig](/docs/transformers/main/id/model_doc/gemma4_unified_assistant#transformers.Gemma4UnifiedAssistantConfig) configuration class: [Gemma4UnifiedAssistantForCausalLM](/docs/transformers/main/id/model_doc/gemma4_unified_assistant#transformers.Gemma4UnifiedAssistantForCausalLM) (Gemma4UnifiedAssistantConfig model) - [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) configuration class: [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (Gemma4UnifiedConfig model) - [Gemma4UnifiedTextConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedTextConfig) configuration class: [Gemma4UnifiedForCausalLM](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForCausalLM) (Gemma4UnifiedTextConfig model) - [GemmaConfig](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaConfig) configuration class: [GemmaForCausalLM](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForCausalLM) (GemmaConfig model) - [GitConfig](/docs/transformers/main/id/model_doc/git#transformers.GitConfig) configuration class: [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (GitConfig model) - [Glm4Config](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Config) configuration class: [Glm4ForCausalLM](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForCausalLM) (Glm4Config model) - [Glm4MoeConfig](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeConfig) configuration class: [Glm4MoeForCausalLM](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeForCausalLM) (Glm4MoeConfig model) - [Glm4MoeLiteConfig](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteConfig) configuration class: [Glm4MoeLiteForCausalLM](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteForCausalLM) (Glm4MoeLiteConfig model) - [GlmConfig](/docs/transformers/main/id/model_doc/glm#transformers.GlmConfig) configuration class: [GlmForCausalLM](/docs/transformers/main/id/model_doc/glm#transformers.GlmForCausalLM) (GlmConfig model) - [GlmMoeDsaConfig](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaConfig) configuration class: [GlmMoeDsaForCausalLM](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaForCausalLM) (GlmMoeDsaConfig model) - [GotOcr2Config](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Config) configuration class: [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (GotOcr2Config model) - [GptOssConfig](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssConfig) configuration class: [GptOssForCausalLM](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForCausalLM) (GptOssConfig model) - [GraniteConfig](/docs/transformers/main/id/model_doc/granite#transformers.GraniteConfig) configuration class: [GraniteForCausalLM](/docs/transformers/main/id/model_doc/granite#transformers.GraniteForCausalLM) (GraniteConfig model) - [GraniteMoeConfig](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeConfig) configuration class: [GraniteMoeForCausalLM](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeForCausalLM) (GraniteMoeConfig model) - [GraniteMoeHybridConfig](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridConfig) configuration class: [GraniteMoeHybridForCausalLM](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridForCausalLM) (GraniteMoeHybridConfig model) - [GraniteMoeSWAConfig](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAConfig) configuration class: [GraniteMoeSWAForCausalLM](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAForCausalLM) (GraniteMoeSWAConfig model) - [GraniteMoeSharedConfig](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedConfig) configuration class: [GraniteMoeSharedForCausalLM](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedForCausalLM) (GraniteMoeSharedConfig model) - [GraniteSWAConfig](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAConfig) configuration class: [GraniteSWAForCausalLM](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAForCausalLM) (GraniteSWAConfig model) - [HYV3Config](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3Config) configuration class: [HYV3ForCausalLM](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3ForCausalLM) (HYV3Config model) - [HYV4Config](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4Config) configuration class: [HYV4ForCausalLM](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4ForCausalLM) (HYV4Config model) - [HeliumConfig](/docs/transformers/main/id/model_doc/helium#transformers.HeliumConfig) configuration class: [HeliumForCausalLM](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForCausalLM) (HeliumConfig model) - [HrmTextConfig](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextConfig) configuration class: [HrmTextForCausalLM](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextForCausalLM) (HrmTextConfig model) - [HunYuanDenseV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Config) configuration class: [HunYuanDenseV1ForCausalLM](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1ForCausalLM) (HunYuanDenseV1Config model) - [HunYuanMoEV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Config) configuration class: [HunYuanMoEV1ForCausalLM](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1ForCausalLM) (HunYuanMoEV1Config model) - [HyperCLOVAXConfig](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXConfig) configuration class: [HyperCLOVAXForCausalLM](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXForCausalLM) (HyperCLOVAXConfig model) - [InklingTextConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingTextConfig) configuration class: [InklingForCausalLM](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForCausalLM) (InklingTextConfig model) - [Jais2Config](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2Config) configuration class: [Jais2ForCausalLM](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2ForCausalLM) (Jais2Config model) - [JambaConfig](/docs/transformers/main/id/model_doc/jamba#transformers.JambaConfig) configuration class: [JambaForCausalLM](/docs/transformers/main/id/model_doc/jamba#transformers.JambaForCausalLM) (JambaConfig model) - [JetMoeConfig](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeConfig) configuration class: [JetMoeForCausalLM](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeForCausalLM) (JetMoeConfig model) - [KimiLinearConfig](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearConfig) configuration class: [KimiLinearForCausalLM](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearForCausalLM) (KimiLinearConfig model) - [Kolibri1Config](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1Config) configuration class: [Kolibri1ForCausalLM](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1ForCausalLM) (Kolibri1Config model) - [LagunaConfig](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaConfig) configuration class: [LagunaForCausalLM](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaForCausalLM) (LagunaConfig model) - [Lfm2Config](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2Config) configuration class: [Lfm2ForCausalLM](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2ForCausalLM) (Lfm2Config model) - [Lfm2MoeConfig](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeConfig) configuration class: [Lfm2MoeForCausalLM](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeForCausalLM) (Lfm2MoeConfig model) - [Lfm3Config](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3Config) configuration class: [Lfm3ForCausalLM](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3ForCausalLM) (Lfm3Config model) - [Llama4Config](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Config) configuration class: [Llama4ForCausalLM](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForCausalLM) (Llama4Config model) - [Llama4TextConfig](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4TextConfig) configuration class: [Llama4ForCausalLM](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForCausalLM) (Llama4TextConfig model) - [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig) configuration class: [LlamaForCausalLM](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaForCausalLM) (LlamaConfig model) - [LongcatFlashConfig](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashConfig) configuration class: [LongcatFlashForCausalLM](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashForCausalLM) (LongcatFlashConfig model) - [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig) configuration class: [MBartForCausalLM](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForCausalLM) (MBartConfig model) - [Mamba2Config](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2Config) configuration class: [Mamba2ForCausalLM](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2ForCausalLM) (Mamba2Config model) - [MambaConfig](/docs/transformers/main/id/model_doc/mamba#transformers.MambaConfig) configuration class: [MambaForCausalLM](/docs/transformers/main/id/model_doc/mamba#transformers.MambaForCausalLM) (MambaConfig model) - [MarianConfig](/docs/transformers/main/id/model_doc/marian#transformers.MarianConfig) configuration class: [MarianForCausalLM](/docs/transformers/main/id/model_doc/marian#transformers.MarianForCausalLM) (MarianConfig model) - [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) configuration class: [MegatronBertForCausalLM](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForCausalLM) (MegatronBertConfig model) - [MellumConfig](/docs/transformers/main/id/model_doc/mellum#transformers.MellumConfig) configuration class: [MellumForCausalLM](/docs/transformers/main/id/model_doc/mellum#transformers.MellumForCausalLM) (MellumConfig model) - [MiMoV2FlashConfig](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashConfig) configuration class: [MiMoV2FlashForCausalLM](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashForCausalLM) (MiMoV2FlashConfig model) - [MiniCPM3Config](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Config) configuration class: [MiniCPM3ForCausalLM](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3ForCausalLM) (MiniCPM3Config model) - [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig) configuration class: [MiniMaxForCausalLM](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForCausalLM) (MiniMaxConfig model) - [MiniMaxM2Config](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2Config) configuration class: [MiniMaxM2ForCausalLM](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2ForCausalLM) (MiniMaxM2Config model) - [MiniMaxM3VLTextConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLTextConfig) configuration class: [MiniMaxM3VLForCausalLM](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLForCausalLM) (MiniMaxM3VLTextConfig model) - [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config) configuration class: [Ministral3ForCausalLM](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForCausalLM) (Ministral3Config model) - [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig) configuration class: [MinistralForCausalLM](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForCausalLM) (MinistralConfig model) - [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig) configuration class: [MistralForCausalLM](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForCausalLM) (MistralConfig model) - [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig) configuration class: [MixtralForCausalLM](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForCausalLM) (MixtralConfig model) - [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) configuration class: [MllamaForCausalLM](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForCausalLM) (MllamaConfig model) - [ModernBertDecoderConfig](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderConfig) configuration class: [ModernBertDecoderForCausalLM](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderForCausalLM) (ModernBertDecoderConfig model) - [MoshiConfig](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiConfig) configuration class: [MoshiForCausalLM](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiForCausalLM) (MoshiConfig model) - [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig) configuration class: [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (MossTranscribeDiarizeConfig model) - [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) configuration class: [MptForCausalLM](/docs/transformers/main/id/model_doc/mpt#transformers.MptForCausalLM) (MptConfig model) - [MusicgenConfig](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenConfig) configuration class: [MusicgenForCausalLM](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenForCausalLM) (MusicgenConfig model) - [MusicgenMelodyConfig](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyConfig) configuration class: [MusicgenMelodyForCausalLM](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyForCausalLM) (MusicgenMelodyConfig model) - [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig) configuration class: [MvpForCausalLM](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForCausalLM) (MvpConfig model) - [NanoChatConfig](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatConfig) configuration class: [NanoChatForCausalLM](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatForCausalLM) (NanoChatConfig model) - [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig) configuration class: [NemotronForCausalLM](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForCausalLM) (NemotronConfig model) - [NemotronHConfig](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHConfig) configuration class: [NemotronHForCausalLM](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHForCausalLM) (NemotronHConfig model) - [OPTConfig](/docs/transformers/main/id/model_doc/opt#transformers.OPTConfig) configuration class: [OPTForCausalLM](/docs/transformers/main/id/model_doc/opt#transformers.OPTForCausalLM) (OPTConfig model) - [Olmo2Config](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Config) configuration class: [Olmo2ForCausalLM](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2ForCausalLM) (Olmo2Config model) - [Olmo3Config](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Config) configuration class: [Olmo3ForCausalLM](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3ForCausalLM) (Olmo3Config model) - [OlmoConfig](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoConfig) configuration class: [OlmoForCausalLM](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoForCausalLM) (OlmoConfig model) - [OlmoHybridConfig](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridConfig) configuration class: [OlmoHybridForCausalLM](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridForCausalLM) (OlmoHybridConfig model) - [OlmoeConfig](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeConfig) configuration class: [OlmoeForCausalLM](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeForCausalLM) (OlmoeConfig model) - [OpenAIGPTConfig](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTConfig) configuration class: [OpenAIGPTLMHeadModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTLMHeadModel) (OpenAIGPTConfig model) - [PLBartConfig](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartConfig) configuration class: [PLBartForCausalLM](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForCausalLM) (PLBartConfig model) - [PegasusConfig](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusConfig) configuration class: [PegasusForCausalLM](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusForCausalLM) (PegasusConfig model) - [PersimmonConfig](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonConfig) configuration class: [PersimmonForCausalLM](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForCausalLM) (PersimmonConfig model) - [Phi3Config](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Config) configuration class: [Phi3ForCausalLM](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForCausalLM) (Phi3Config model) - [Phi4MultimodalConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalConfig) configuration class: [Phi4MultimodalForCausalLM](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalForCausalLM) (Phi4MultimodalConfig model) - [PhiConfig](/docs/transformers/main/id/model_doc/phi#transformers.PhiConfig) configuration class: [PhiForCausalLM](/docs/transformers/main/id/model_doc/phi#transformers.PhiForCausalLM) (PhiConfig model) - [PhimoeConfig](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeConfig) configuration class: [PhimoeForCausalLM](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeForCausalLM) (PhimoeConfig model) - [ProphetNetConfig](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetConfig) configuration class: [ProphetNetForCausalLM](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetForCausalLM) (ProphetNetConfig model) - [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config) configuration class: [Qwen2ForCausalLM](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForCausalLM) (Qwen2Config model) - [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig) configuration class: [Qwen2MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForCausalLM) (Qwen2MoeConfig model) - [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config) configuration class: [Qwen3ForCausalLM](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForCausalLM) (Qwen3Config model) - [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig) configuration class: [Qwen3MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForCausalLM) (Qwen3MoeConfig model) - [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig) configuration class: [Qwen3NextForCausalLM](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForCausalLM) (Qwen3NextConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5ForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForCausalLM) (Qwen3_5Config model) - [Qwen3_5MoeConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeConfig) configuration class: [Qwen3_5MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForCausalLM) (Qwen3_5MoeConfig model) - [Qwen3_5MoeTextConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeTextConfig) configuration class: [Qwen3_5MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForCausalLM) (Qwen3_5MoeTextConfig model) - [Qwen3_5TextConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextConfig) configuration class: [Qwen3_5ForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForCausalLM) (Qwen3_5TextConfig model) - [Qwen4ExpConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpConfig) configuration class: [Qwen4ExpForCausalLM](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForCausalLM) (Qwen4ExpConfig model) - [Qwen4ExpTextConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpTextConfig) configuration class: [Qwen4ExpForCausalLM](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForCausalLM) (Qwen4ExpTextConfig model) - [RecurrentGemmaConfig](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaConfig) configuration class: [RecurrentGemmaForCausalLM](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaForCausalLM) (RecurrentGemmaConfig model) - [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig) configuration class: [ReformerModelWithLMHead](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerModelWithLMHead) (ReformerConfig model) - [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig) configuration class: [RemBertForCausalLM](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForCausalLM) (RemBertConfig model) - [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) configuration class: [RoCBertForCausalLM](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForCausalLM) (RoCBertConfig model) - [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig) configuration class: [RoFormerForCausalLM](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForCausalLM) (RoFormerConfig model) - [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) configuration class: [RobertaForCausalLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForCausalLM) (RobertaConfig model) - [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) configuration class: [RobertaPreLayerNormForCausalLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForCausalLM) (RobertaPreLayerNormConfig model) - [RwkvConfig](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvConfig) configuration class: [RwkvForCausalLM](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvForCausalLM) (RwkvConfig model) - [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig) configuration class: [SeedOssForCausalLM](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForCausalLM) (SeedOssConfig model) - [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config) configuration class: [SmolLM3ForCausalLM](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForCausalLM) (SmolLM3Config model) - [SolarOpenConfig](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenConfig) configuration class: [SolarOpenForCausalLM](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenForCausalLM) (SolarOpenConfig model) - [StableLmConfig](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmConfig) configuration class: [StableLmForCausalLM](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForCausalLM) (StableLmConfig model) - [Starcoder2Config](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Config) configuration class: [Starcoder2ForCausalLM](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForCausalLM) (Starcoder2Config model) - [TrOCRConfig](/docs/transformers/main/id/model_doc/trocr#transformers.TrOCRConfig) configuration class: [TrOCRForCausalLM](/docs/transformers/main/id/model_doc/trocr#transformers.TrOCRForCausalLM) (TrOCRConfig model) - [VaultGemmaConfig](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaConfig) configuration class: [VaultGemmaForCausalLM](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaForCausalLM) (VaultGemmaConfig model) - [WhisperConfig](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperConfig) configuration class: [WhisperForCausalLM](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForCausalLM) (WhisperConfig model) - [XGLMConfig](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMConfig) configuration class: [XGLMForCausalLM](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMForCausalLM) (XGLMConfig model) - [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) configuration class: [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (XLMConfig model) - [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) configuration class: [XLMRobertaForCausalLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForCausalLM) (XLMRobertaConfig model) - [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) configuration class: [XLMRobertaXLForCausalLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForCausalLM) (XLMRobertaXLConfig model) - [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) configuration class: [XLNetLMHeadModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetLMHeadModel) (XLNetConfig model) - [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) configuration class: [XmodForCausalLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForCausalLM) (XmodConfig model) - [YoutuConfig](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuConfig) configuration class: [YoutuForCausalLM](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuForCausalLM) (YoutuConfig model) - [Zamba2Config](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Config) configuration class: [Zamba2ForCausalLM](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2ForCausalLM) (Zamba2Config model) - [ZambaConfig](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaConfig) configuration class: [ZambaForCausalLM](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaForCausalLM) (ZambaConfig model) - [ZayaConfig](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaConfig) configuration class: [ZayaForCausalLM](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaForCausalLM) (ZayaConfig model) - [xLSTMConfig](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMConfig) configuration class: [xLSTMForCausalLM](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMForCausalLM) (xLSTMConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa kausal) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForCausalLM

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForCausalLM.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForCausalLM.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa kausal) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **afmoe** -- [AfmoeForCausalLM](/docs/transformers/main/id/model_doc/afmoe#transformers.AfmoeForCausalLM) (model AfmoeConfig)
- **apertus** -- [ApertusForCausalLM](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusForCausalLM) (model ApertusConfig)
- **arcee** -- [ArceeForCausalLM](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForCausalLM) (model ArceeConfig)
- **aria_text** -- [AriaTextForCausalLM](/docs/transformers/main/id/model_doc/aria#transformers.AriaTextForCausalLM) (model AriaTextConfig)
- **axk1** -- [AXK1ForCausalLM](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForCausalLM) (model AXK1Config)
- **axk2** -- [AXK2ForCausalLM](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForCausalLM) (model AXK2Config)
- **bailing_hybrid** -- [BailingHybridForCausalLM](/docs/transformers/main/id/model_doc/bailing_hybrid#transformers.BailingHybridForCausalLM) (model BailingHybridConfig)
- **bamba** -- [BambaForCausalLM](/docs/transformers/main/id/model_doc/bamba#transformers.BambaForCausalLM) (model BambaConfig)
- **bart** -- [BartForCausalLM](/docs/transformers/main/id/model_doc/bart#transformers.BartForCausalLM) (model BartConfig)
- **beam** -- [BeamForCausalLM](/docs/transformers/main/id/model_doc/beam#transformers.BeamForCausalLM) (model BeamConfig)
- **bert** -- [BertLMHeadModel](/docs/transformers/main/id/model_doc/bert#transformers.BertLMHeadModel) (model BertConfig)
- **bert-generation** -- [BertGenerationDecoder](/docs/transformers/main/id/model_doc/bert-generation#transformers.BertGenerationDecoder) (model BertGenerationConfig)
- **big_bird** -- [BigBirdForCausalLM](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForCausalLM) (model BigBirdConfig)
- **bigbird_pegasus** -- [BigBirdPegasusForCausalLM](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForCausalLM) (model BigBirdPegasusConfig)
- **biogpt** -- [BioGptForCausalLM](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForCausalLM) (model BioGptConfig)
- **bitnet** -- [BitNetForCausalLM](/docs/transformers/main/id/model_doc/bitnet#transformers.BitNetForCausalLM) (model BitNetConfig)
- **blenderbot** -- [BlenderbotForCausalLM](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotForCausalLM) (model BlenderbotConfig)
- **blenderbot-small** -- [BlenderbotSmallForCausalLM](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallForCausalLM) (model BlenderbotSmallConfig)
- **bloom** -- [BloomForCausalLM](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForCausalLM) (model BloomConfig)
- **blt** -- [BltForCausalLM](/docs/transformers/main/id/model_doc/blt#transformers.BltForCausalLM) (model BltConfig)
- **camembert** -- [CamembertForCausalLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForCausalLM) (model CamembertConfig)
- **codegen** -- [CodeGenForCausalLM](/docs/transformers/main/id/model_doc/codegen#transformers.CodeGenForCausalLM) (model CodeGenConfig)
- **cohere** -- [CohereForCausalLM](/docs/transformers/main/id/model_doc/cohere#transformers.CohereForCausalLM) (model CohereConfig)
- **cohere2** -- [Cohere2ForCausalLM](/docs/transformers/main/id/model_doc/cohere2#transformers.Cohere2ForCausalLM) (model Cohere2Config)
- **cohere2_moe** -- [Cohere2MoeForCausalLM](/docs/transformers/main/id/model_doc/cohere2_moe#transformers.Cohere2MoeForCausalLM) (model Cohere2MoeConfig)
- **cohere_compass_text** -- [CohereCompassForCausalLM](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForCausalLM) (model CohereCompassTextConfig)
- **cpmant** -- [CpmAntForCausalLM](/docs/transformers/main/id/model_doc/cpmant#transformers.CpmAntForCausalLM) (model CpmAntConfig)
- **ctrl** -- [CTRLLMHeadModel](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLLMHeadModel) (model CTRLConfig)
- **cwm** -- [CwmForCausalLM](/docs/transformers/main/id/model_doc/cwm#transformers.CwmForCausalLM) (model CwmConfig)
- **data2vec-text** -- [Data2VecTextForCausalLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForCausalLM) (model Data2VecTextConfig)
- **dbrx** -- [DbrxForCausalLM](/docs/transformers/main/id/model_doc/dbrx#transformers.DbrxForCausalLM) (model DbrxConfig)
- **deepseek_v2** -- [DeepseekV2ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2ForCausalLM) (model DeepseekV2Config)
- **deepseek_v3** -- [DeepseekV3ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForCausalLM) (model DeepseekV3Config)
- **deepseek_v32** -- [DeepseekV32ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v32#transformers.DeepseekV32ForCausalLM) (model DeepseekV32Config)
- **deepseek_v4** -- [DeepseekV4ForCausalLM](/docs/transformers/main/id/model_doc/deepseek_v4#transformers.DeepseekV4ForCausalLM) (model DeepseekV4Config)
- **diffllama** -- [DiffLlamaForCausalLM](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForCausalLM) (model DiffLlamaConfig)
- **doge** -- [DogeForCausalLM](/docs/transformers/main/id/model_doc/doge#transformers.DogeForCausalLM) (model DogeConfig)
- **dots1** -- [Dots1ForCausalLM](/docs/transformers/main/id/model_doc/dots1#transformers.Dots1ForCausalLM) (model Dots1Config)
- **electra** -- [ElectraForCausalLM](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForCausalLM) (model ElectraConfig)
- **emu3** -- [Emu3ForCausalLM](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForCausalLM) (model Emu3Config)
- **ernie** -- [ErnieForCausalLM](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForCausalLM) (model ErnieConfig)
- **ernie4_5** -- [Ernie4_5ForCausalLM](/docs/transformers/main/id/model_doc/ernie4_5#transformers.Ernie4_5ForCausalLM) (model Ernie4_5Config)
- **ernie4_5_moe** -- [Ernie4_5_MoeForCausalLM](/docs/transformers/main/id/model_doc/ernie4_5_moe#transformers.Ernie4_5_MoeForCausalLM) (model Ernie4_5_MoeConfig)
- **exaone4** -- [Exaone4ForCausalLM](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForCausalLM) (model Exaone4Config)
- **exaone_moe** -- [ExaoneMoeForCausalLM](/docs/transformers/main/id/model_doc/exaone_moe#transformers.ExaoneMoeForCausalLM) (model ExaoneMoeConfig)
- **falcon** -- [FalconForCausalLM](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForCausalLM) (model FalconConfig)
- **falcon_h1** -- [FalconH1ForCausalLM](/docs/transformers/main/id/model_doc/falcon_h1#transformers.FalconH1ForCausalLM) (model FalconH1Config)
- **falcon_mamba** -- [FalconMambaForCausalLM](/docs/transformers/main/id/model_doc/falcon_mamba#transformers.FalconMambaForCausalLM) (model FalconMambaConfig)
- **flex_olmo** -- [FlexOlmoForCausalLM](/docs/transformers/main/id/model_doc/flex_olmo#transformers.FlexOlmoForCausalLM) (model FlexOlmoConfig)
- **flint** -- [FlintForCausalLM](/docs/transformers/main/id/model_doc/flint#transformers.FlintForCausalLM) (model FlintConfig)
- **fuyu** -- [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (model FuyuConfig)
- **gemma** -- [GemmaForCausalLM](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForCausalLM) (model GemmaConfig)
- **gemma2** -- [Gemma2ForCausalLM](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForCausalLM) (model Gemma2Config)
- **gemma3** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model Gemma3Config)
- **gemma3_text** -- [Gemma3ForCausalLM](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForCausalLM) (model Gemma3TextConfig)
- **gemma3n** -- [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (model Gemma3nConfig)
- **gemma3n_text** -- [Gemma3nForCausalLM](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForCausalLM) (model Gemma3nTextConfig)
- **gemma4** -- [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (model Gemma4Config)
- **gemma4_assistant** -- [Gemma4AssistantForCausalLM](/docs/transformers/main/id/model_doc/gemma4_assistant#transformers.Gemma4AssistantForCausalLM) (model Gemma4AssistantConfig)
- **gemma4_text** -- [Gemma4ForCausalLM](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForCausalLM) (model Gemma4TextConfig)
- **gemma4_unified** -- [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (model Gemma4UnifiedConfig)
- **gemma4_unified_assistant** -- [Gemma4UnifiedAssistantForCausalLM](/docs/transformers/main/id/model_doc/gemma4_unified_assistant#transformers.Gemma4UnifiedAssistantForCausalLM) (model Gemma4UnifiedAssistantConfig)
- **gemma4_unified_text** -- [Gemma4UnifiedForCausalLM](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForCausalLM) (model Gemma4UnifiedTextConfig)
- **git** -- [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (model GitConfig)
- **glm** -- [GlmForCausalLM](/docs/transformers/main/id/model_doc/glm#transformers.GlmForCausalLM) (model GlmConfig)
- **glm4** -- [Glm4ForCausalLM](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForCausalLM) (model Glm4Config)
- **glm4_moe** -- [Glm4MoeForCausalLM](/docs/transformers/main/id/model_doc/glm4_moe#transformers.Glm4MoeForCausalLM) (model Glm4MoeConfig)
- **glm4_moe_lite** -- [Glm4MoeLiteForCausalLM](/docs/transformers/main/id/model_doc/glm4_moe_lite#transformers.Glm4MoeLiteForCausalLM) (model Glm4MoeLiteConfig)
- **glm_moe_dsa** -- [GlmMoeDsaForCausalLM](/docs/transformers/main/id/model_doc/glm_moe_dsa#transformers.GlmMoeDsaForCausalLM) (model GlmMoeDsaConfig)
- **got_ocr2** -- [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (model GotOcr2Config)
- **gpt-sw3** -- [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (model GPT2Config)
- **gpt2** -- [GPT2LMHeadModel](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2LMHeadModel) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeForCausalLM](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForCausalLM) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPTNeoForCausalLM](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForCausalLM) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXForCausalLM](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForCausalLM) (model GPTNeoXConfig)
- **gpt_neox_japanese** -- [GPTNeoXJapaneseForCausalLM](/docs/transformers/main/id/model_doc/gpt_neox_japanese#transformers.GPTNeoXJapaneseForCausalLM) (model GPTNeoXJapaneseConfig)
- **gpt_oss** -- [GptOssForCausalLM](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForCausalLM) (model GptOssConfig)
- **gptj** -- [GPTJForCausalLM](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForCausalLM) (model GPTJConfig)
- **granite** -- [GraniteForCausalLM](/docs/transformers/main/id/model_doc/granite#transformers.GraniteForCausalLM) (model GraniteConfig)
- **granite_swa** -- [GraniteSWAForCausalLM](/docs/transformers/main/id/model_doc/granite_swa#transformers.GraniteSWAForCausalLM) (model GraniteSWAConfig)
- **granitemoe** -- [GraniteMoeForCausalLM](/docs/transformers/main/id/model_doc/granitemoe#transformers.GraniteMoeForCausalLM) (model GraniteMoeConfig)
- **granitemoe_swa** -- [GraniteMoeSWAForCausalLM](/docs/transformers/main/id/model_doc/granitemoe_swa#transformers.GraniteMoeSWAForCausalLM) (model GraniteMoeSWAConfig)
- **granitemoehybrid** -- [GraniteMoeHybridForCausalLM](/docs/transformers/main/id/model_doc/granitemoehybrid#transformers.GraniteMoeHybridForCausalLM) (model GraniteMoeHybridConfig)
- **granitemoeshared** -- [GraniteMoeSharedForCausalLM](/docs/transformers/main/id/model_doc/granitemoeshared#transformers.GraniteMoeSharedForCausalLM) (model GraniteMoeSharedConfig)
- **helium** -- [HeliumForCausalLM](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForCausalLM) (model HeliumConfig)
- **hrm_text** -- [HrmTextForCausalLM](/docs/transformers/main/id/model_doc/hrm_text#transformers.HrmTextForCausalLM) (model HrmTextConfig)
- **hunyuan_v1_dense** -- [HunYuanDenseV1ForCausalLM](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1ForCausalLM) (model HunYuanDenseV1Config)
- **hunyuan_v1_moe** -- [HunYuanMoEV1ForCausalLM](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1ForCausalLM) (model HunYuanMoEV1Config)
- **hy_v3** -- [HYV3ForCausalLM](/docs/transformers/main/id/model_doc/hy_v3#transformers.HYV3ForCausalLM) (model HYV3Config)
- **hy_v4** -- [HYV4ForCausalLM](/docs/transformers/main/id/model_doc/hy_v4#transformers.HYV4ForCausalLM) (model HYV4Config)
- **hyperclovax** -- [HyperCLOVAXForCausalLM](/docs/transformers/main/id/model_doc/hyperclovax#transformers.HyperCLOVAXForCausalLM) (model HyperCLOVAXConfig)
- **inkling_text** -- [InklingForCausalLM](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForCausalLM) (model InklingTextConfig)
- **jais2** -- [Jais2ForCausalLM](/docs/transformers/main/id/model_doc/jais2#transformers.Jais2ForCausalLM) (model Jais2Config)
- **jamba** -- [JambaForCausalLM](/docs/transformers/main/id/model_doc/jamba#transformers.JambaForCausalLM) (model JambaConfig)
- **jetmoe** -- [JetMoeForCausalLM](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeForCausalLM) (model JetMoeConfig)
- **kimi_linear** -- [KimiLinearForCausalLM](/docs/transformers/main/id/model_doc/kimi_linear#transformers.KimiLinearForCausalLM) (model KimiLinearConfig)
- **kolibri1** -- [Kolibri1ForCausalLM](/docs/transformers/main/id/model_doc/kolibri1#transformers.Kolibri1ForCausalLM) (model Kolibri1Config)
- **laguna** -- [LagunaForCausalLM](/docs/transformers/main/id/model_doc/laguna#transformers.LagunaForCausalLM) (model LagunaConfig)
- **lfm2** -- [Lfm2ForCausalLM](/docs/transformers/main/id/model_doc/lfm2#transformers.Lfm2ForCausalLM) (model Lfm2Config)
- **lfm2_moe** -- [Lfm2MoeForCausalLM](/docs/transformers/main/id/model_doc/lfm2_moe#transformers.Lfm2MoeForCausalLM) (model Lfm2MoeConfig)
- **lfm3** -- [Lfm3ForCausalLM](/docs/transformers/main/id/model_doc/lfm3#transformers.Lfm3ForCausalLM) (model Lfm3Config)
- **llama** -- [LlamaForCausalLM](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaForCausalLM) (model LlamaConfig)
- **llama4** -- [Llama4ForCausalLM](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForCausalLM) (model Llama4Config)
- **llama4_text** -- [Llama4ForCausalLM](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForCausalLM) (model Llama4TextConfig)
- **longcat_flash** -- [LongcatFlashForCausalLM](/docs/transformers/main/id/model_doc/longcat_flash#transformers.LongcatFlashForCausalLM) (model LongcatFlashConfig)
- **mamba** -- [MambaForCausalLM](/docs/transformers/main/id/model_doc/mamba#transformers.MambaForCausalLM) (model MambaConfig)
- **mamba2** -- [Mamba2ForCausalLM](/docs/transformers/main/id/model_doc/mamba2#transformers.Mamba2ForCausalLM) (model Mamba2Config)
- **marian** -- [MarianForCausalLM](/docs/transformers/main/id/model_doc/marian#transformers.MarianForCausalLM) (model MarianConfig)
- **mbart** -- [MBartForCausalLM](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForCausalLM) (model MBartConfig)
- **megatron-bert** -- [MegatronBertForCausalLM](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForCausalLM) (model MegatronBertConfig)
- **mellum** -- [MellumForCausalLM](/docs/transformers/main/id/model_doc/mellum#transformers.MellumForCausalLM) (model MellumConfig)
- **mimo_v2_flash** -- [MiMoV2FlashForCausalLM](/docs/transformers/main/id/model_doc/mimo_v2_flash#transformers.MiMoV2FlashForCausalLM) (model MiMoV2FlashConfig)
- **minicpm3** -- [MiniCPM3ForCausalLM](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3ForCausalLM) (model MiniCPM3Config)
- **minimax** -- [MiniMaxForCausalLM](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForCausalLM) (model MiniMaxConfig)
- **minimax_m2** -- [MiniMaxM2ForCausalLM](/docs/transformers/main/id/model_doc/minimax_m2#transformers.MiniMaxM2ForCausalLM) (model MiniMaxM2Config)
- **minimax_m3_vl_text** -- [MiniMaxM3VLForCausalLM](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLForCausalLM) (model MiniMaxM3VLTextConfig)
- **ministral** -- [MinistralForCausalLM](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForCausalLM) (model MinistralConfig)
- **ministral3** -- [Ministral3ForCausalLM](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForCausalLM) (model Ministral3Config)
- **mistral** -- [MistralForCausalLM](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForCausalLM) (model MistralConfig)
- **mixtral** -- [MixtralForCausalLM](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForCausalLM) (model MixtralConfig)
- **mllama** -- [MllamaForCausalLM](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForCausalLM) (model MllamaConfig)
- **modernbert-decoder** -- [ModernBertDecoderForCausalLM](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderForCausalLM) (model ModernBertDecoderConfig)
- **moshi** -- [MoshiForCausalLM](/docs/transformers/main/id/model_doc/moshi#transformers.MoshiForCausalLM) (model MoshiConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (model MossTranscribeDiarizeConfig)
- **mpt** -- [MptForCausalLM](/docs/transformers/main/id/model_doc/mpt#transformers.MptForCausalLM) (model MptConfig)
- **musicgen** -- [MusicgenForCausalLM](/docs/transformers/main/id/model_doc/musicgen#transformers.MusicgenForCausalLM) (model MusicgenConfig)
- **musicgen_melody** -- [MusicgenMelodyForCausalLM](/docs/transformers/main/id/model_doc/musicgen_melody#transformers.MusicgenMelodyForCausalLM) (model MusicgenMelodyConfig)
- **mvp** -- [MvpForCausalLM](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForCausalLM) (model MvpConfig)
- **nanochat** -- [NanoChatForCausalLM](/docs/transformers/main/id/model_doc/nanochat#transformers.NanoChatForCausalLM) (model NanoChatConfig)
- **nemotron** -- [NemotronForCausalLM](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForCausalLM) (model NemotronConfig)
- **nemotron_h** -- [NemotronHForCausalLM](/docs/transformers/main/id/model_doc/nemotron_h#transformers.NemotronHForCausalLM) (model NemotronHConfig)
- **olmo** -- [OlmoForCausalLM](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoForCausalLM) (model OlmoConfig)
- **olmo2** -- [Olmo2ForCausalLM](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2ForCausalLM) (model Olmo2Config)
- **olmo3** -- [Olmo3ForCausalLM](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3ForCausalLM) (model Olmo3Config)
- **olmo_hybrid** -- [OlmoHybridForCausalLM](/docs/transformers/main/id/model_doc/olmo_hybrid#transformers.OlmoHybridForCausalLM) (model OlmoHybridConfig)
- **olmoe** -- [OlmoeForCausalLM](/docs/transformers/main/id/model_doc/olmoe#transformers.OlmoeForCausalLM) (model OlmoeConfig)
- **openai-gpt** -- [OpenAIGPTLMHeadModel](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTLMHeadModel) (model OpenAIGPTConfig)
- **opt** -- [OPTForCausalLM](/docs/transformers/main/id/model_doc/opt#transformers.OPTForCausalLM) (model OPTConfig)
- **pegasus** -- [PegasusForCausalLM](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusForCausalLM) (model PegasusConfig)
- **persimmon** -- [PersimmonForCausalLM](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForCausalLM) (model PersimmonConfig)
- **phi** -- [PhiForCausalLM](/docs/transformers/main/id/model_doc/phi#transformers.PhiForCausalLM) (model PhiConfig)
- **phi3** -- [Phi3ForCausalLM](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForCausalLM) (model Phi3Config)
- **phi4_multimodal** -- [Phi4MultimodalForCausalLM](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalForCausalLM) (model Phi4MultimodalConfig)
- **phimoe** -- [PhimoeForCausalLM](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeForCausalLM) (model PhimoeConfig)
- **plbart** -- [PLBartForCausalLM](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForCausalLM) (model PLBartConfig)
- **prophetnet** -- [ProphetNetForCausalLM](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetForCausalLM) (model ProphetNetConfig)
- **qwen2** -- [Qwen2ForCausalLM](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForCausalLM) (model Qwen2Config)
- **qwen2_moe** -- [Qwen2MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForCausalLM) (model Qwen2MoeConfig)
- **qwen3** -- [Qwen3ForCausalLM](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForCausalLM) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5ForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForCausalLM) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForCausalLM) (model Qwen3_5MoeConfig)
- **qwen3_5_moe_text** -- [Qwen3_5MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForCausalLM) (model Qwen3_5MoeTextConfig)
- **qwen3_5_text** -- [Qwen3_5ForCausalLM](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForCausalLM) (model Qwen3_5TextConfig)
- **qwen3_moe** -- [Qwen3MoeForCausalLM](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForCausalLM) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextForCausalLM](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForCausalLM) (model Qwen3NextConfig)
- **qwen4_exp** -- [Qwen4ExpForCausalLM](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForCausalLM) (model Qwen4ExpConfig)
- **qwen4_exp_text** -- [Qwen4ExpForCausalLM](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForCausalLM) (model Qwen4ExpTextConfig)
- **recurrent_gemma** -- [RecurrentGemmaForCausalLM](/docs/transformers/main/id/model_doc/recurrent_gemma#transformers.RecurrentGemmaForCausalLM) (model RecurrentGemmaConfig)
- **reformer** -- [ReformerModelWithLMHead](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerModelWithLMHead) (model ReformerConfig)
- **rembert** -- [RemBertForCausalLM](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForCausalLM) (model RemBertConfig)
- **roberta** -- [RobertaForCausalLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForCausalLM) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForCausalLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForCausalLM) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForCausalLM](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForCausalLM) (model RoCBertConfig)
- **roformer** -- [RoFormerForCausalLM](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForCausalLM) (model RoFormerConfig)
- **rwkv** -- [RwkvForCausalLM](/docs/transformers/main/id/model_doc/rwkv#transformers.RwkvForCausalLM) (model RwkvConfig)
- **seed_oss** -- [SeedOssForCausalLM](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForCausalLM) (model SeedOssConfig)
- **smollm3** -- [SmolLM3ForCausalLM](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForCausalLM) (model SmolLM3Config)
- **solar_open** -- [SolarOpenForCausalLM](/docs/transformers/main/id/model_doc/solar_open#transformers.SolarOpenForCausalLM) (model SolarOpenConfig)
- **stablelm** -- [StableLmForCausalLM](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForCausalLM) (model StableLmConfig)
- **starcoder2** -- [Starcoder2ForCausalLM](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForCausalLM) (model Starcoder2Config)
- **trocr** -- [TrOCRForCausalLM](/docs/transformers/main/id/model_doc/trocr#transformers.TrOCRForCausalLM) (model TrOCRConfig)
- **vaultgemma** -- [VaultGemmaForCausalLM](/docs/transformers/main/id/model_doc/vaultgemma#transformers.VaultGemmaForCausalLM) (model VaultGemmaConfig)
- **whisper** -- [WhisperForCausalLM](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForCausalLM) (model WhisperConfig)
- **xglm** -- [XGLMForCausalLM](/docs/transformers/main/id/model_doc/xglm#transformers.XGLMForCausalLM) (model XGLMConfig)
- **xlm** -- [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForCausalLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForCausalLM) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForCausalLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForCausalLM) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetLMHeadModel](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetLMHeadModel) (model XLNetConfig)
- **xlstm** -- [xLSTMForCausalLM](/docs/transformers/main/id/model_doc/xlstm#transformers.xLSTMForCausalLM) (model xLSTMConfig)
- **xmod** -- [XmodForCausalLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForCausalLM) (model XmodConfig)
- **youtu** -- [YoutuForCausalLM](/docs/transformers/main/id/model_doc/youtu#transformers.YoutuForCausalLM) (model YoutuConfig)
- **zamba** -- [ZambaForCausalLM](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaForCausalLM) (model ZambaConfig)
- **zamba2** -- [Zamba2ForCausalLM](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2ForCausalLM) (model Zamba2Config)
- **zaya** -- [ZayaForCausalLM](/docs/transformers/main/id/model_doc/zaya#transformers.ZayaForCausalLM) (model ZayaConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForCausalLM

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForCausalLM.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForCausalLM.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForMaskedLM[[transformers.AutoModelForMaskedLM]]

#### transformers.AutoModelForMaskedLM[[transformers.AutoModelForMaskedLM]]

```python
transformers.AutoModelForMaskedLM(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2328)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berpenutup) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForMaskedLM.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig): [AlbertForMaskedLM](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForMaskedLM) (model AlbertConfig) - Kelas konfigurasi [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig): [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (model BartConfig) - Kelas konfigurasi [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig): [BertForMaskedLM](/docs/transformers/main/id/model_doc/bert#transformers.BertForMaskedLM) (model BertConfig) - Kelas konfigurasi [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig): [BigBirdForMaskedLM](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForMaskedLM) (model BigBirdConfig) - Kelas konfigurasi [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig): [CamembertForMaskedLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMaskedLM) (model CamembertConfig) - Kelas konfigurasi [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig): [ConvBertForMaskedLM](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForMaskedLM) (model ConvBertConfig) - Kelas konfigurasi [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig): [Data2VecTextForMaskedLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMaskedLM) (model Data2VecTextConfig) - Kelas konfigurasi [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig): [DebertaForMaskedLM](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForMaskedLM) (model DebertaConfig) - Kelas konfigurasi [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config): [DebertaV2ForMaskedLM](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMaskedLM) (model DebertaV2Config) - Kelas konfigurasi [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig): [DistilBertForMaskedLM](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMaskedLM) (model DistilBertConfig) - Kelas konfigurasi [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig): [ElectraForMaskedLM](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForMaskedLM) (model ElectraConfig) - Kelas konfigurasi [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig): [ErnieForMaskedLM](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForMaskedLM) (model ErnieConfig) - Kelas konfigurasi [EsmConfig](/docs/transformers/main/id/model_doc/esm#transformers.EsmConfig): [EsmForMaskedLM](/docs/transformers/main/id/model_doc/esm#transformers.EsmForMaskedLM) (model EsmConfig) - Kelas konfigurasi [EsmcConfig](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcConfig): [EsmcForMaskedLM](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForMaskedLM) (model EsmcConfig) - Kelas konfigurasi [EuroBertConfig](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertConfig): [EuroBertForMaskedLM](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForMaskedLM) (model EuroBertConfig) - Kelas konfigurasi [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig): [FNetForMaskedLM](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForMaskedLM) (model FNetConfig) - Kelas konfigurasi [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig): [FlaubertWithLMHeadModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertWithLMHeadModel) (model FlaubertConfig) - Kelas konfigurasi [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig): [FunnelForMaskedLM](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForMaskedLM) (model FunnelConfig) - Kelas konfigurasi [GteConfig](/docs/transformers/main/id/model_doc/gte#transformers.GteConfig): [GteForMaskedLM](/docs/transformers/main/id/model_doc/gte#transformers.GteForMaskedLM) (model GteConfig) - Kelas konfigurasi [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig): [IBertForMaskedLM](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMaskedLM) (model IBertConfig) - Kelas konfigurasi [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config): [JinaEmbeddingsV3ForMaskedLM](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForMaskedLM) (model JinaEmbeddingsV3Config) - Kelas konfigurasi [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig): [LayoutLMForMaskedLM](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForMaskedLM) (model LayoutLMConfig) - Kelas konfigurasi [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig): [LongformerForMaskedLM](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMaskedLM) (model LongformerConfig) - Kelas konfigurasi [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig): [LukeForMaskedLM](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMaskedLM) (model LukeConfig) - Kelas konfigurasi [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig): [MBartForConditionalGeneration](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForConditionalGeneration) (model MBartConfig) - Kelas konfigurasi [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig): [MPNetForMaskedLM](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMaskedLM) (model MPNetConfig) - Kelas konfigurasi [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig): [MegatronBertForMaskedLM](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForMaskedLM) (model MegatronBertConfig) - Kelas konfigurasi [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig): [MobileBertForMaskedLM](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForMaskedLM) (model MobileBertConfig) - Kelas konfigurasi [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig): [ModernBertForMaskedLM](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForMaskedLM) (model ModernBertConfig) - Kelas konfigurasi [ModernVBertConfig](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertConfig): [ModernVBertForMaskedLM](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForMaskedLM) (model ModernVBertConfig) - Kelas konfigurasi [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig): [MraForMaskedLM](/docs/transformers/main/id/model_doc/mra#transformers.MraForMaskedLM) (model MraConfig) - Kelas konfigurasi [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig): [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (model MvpConfig) - Kelas konfigurasi [NeoMMEConfig](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEConfig): [NeoMMEForMaskedLM](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEForMaskedLM) (model NeoMMEConfig) - Kelas konfigurasi [NomicBertConfig](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertConfig): [NomicBertForMaskedLM](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForMaskedLM) (model NomicBertConfig) - Kelas konfigurasi [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig): [NystromformerForMaskedLM](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForMaskedLM) (model NystromformerConfig) - Kelas konfigurasi [PerceiverConfig](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverConfig): [PerceiverForMaskedLM](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForMaskedLM) (model PerceiverConfig) - Kelas konfigurasi [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig): [ReformerForMaskedLM](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForMaskedLM) (model ReformerConfig) - Kelas konfigurasi [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig): [RemBertForMaskedLM](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForMaskedLM) (model RemBertConfig) - Kelas konfigurasi [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig): [RoCBertForMaskedLM](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForMaskedLM) (model RoCBertConfig) - Kelas konfigurasi [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig): [RoFormerForMaskedLM](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForMaskedLM) (model RoFormerConfig) - Kelas konfigurasi [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig): [RobertaForMaskedLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMaskedLM) (model RobertaConfig) - Kelas konfigurasi [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig): [RobertaPreLayerNormForMaskedLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMaskedLM) (model RobertaPreLayerNormConfig) - Kelas konfigurasi [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig): [SqueezeBertForMaskedLM](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMaskedLM) (model SqueezeBertConfig) - Kelas konfigurasi [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig): [TapasForMaskedLM](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForMaskedLM) (model TapasConfig) - Kelas konfigurasi [UltraVBertConfig](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertConfig): [UltraVBertForMaskedLM](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForMaskedLM) (model UltraVBertConfig) - Kelas konfigurasi [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig): [UltrabertForMaskedLM](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForMaskedLM) (model UltrabertConfig) - Kelas konfigurasi [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig): [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (model XLMConfig) - Kelas konfigurasi [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig): [XLMRobertaForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMaskedLM) (model XLMRobertaConfig) - Kelas konfigurasi [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig): [XLMRobertaXLForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMaskedLM) (model XLMRobertaXLConfig) - Kelas konfigurasi [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig): [XmodForMaskedLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMaskedLM) (model XmodConfig) - Kelas konfigurasi [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig): [YosoForMaskedLM](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForMaskedLM) (model YosoConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berpenutup) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMaskedLM

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForMaskedLM.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForMaskedLM.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berpenyamar) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForMaskedLM](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForMaskedLM) (model AlbertConfig)
- **bart** -- [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (model BartConfig)
- **bert** -- [BertForMaskedLM](/docs/transformers/main/id/model_doc/bert#transformers.BertForMaskedLM) (model BertConfig)
- **big_bird** -- [BigBirdForMaskedLM](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForMaskedLM) (model BigBirdConfig)
- **camembert** -- [CamembertForMaskedLM](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMaskedLM) (model CamembertConfig)
- **convbert** -- [ConvBertForMaskedLM](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForMaskedLM) (model ConvBertConfig)
- **data2vec-text** -- [Data2VecTextForMaskedLM](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMaskedLM) (model Data2VecTextConfig)
- **deberta** -- [DebertaForMaskedLM](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForMaskedLM) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2ForMaskedLM](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMaskedLM) (model DebertaV2Config)
- **distilbert** -- [DistilBertForMaskedLM](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMaskedLM) (model DistilBertConfig)
- **electra** -- [ElectraForMaskedLM](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForMaskedLM) (model ElectraConfig)
- **ernie** -- [ErnieForMaskedLM](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForMaskedLM) (model ErnieConfig)
- **esm** -- [EsmForMaskedLM](/docs/transformers/main/id/model_doc/esm#transformers.EsmForMaskedLM) (model EsmConfig)
- **esmc** -- [EsmcForMaskedLM](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForMaskedLM) (model EsmcConfig)
- **eurobert** -- [EuroBertForMaskedLM](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForMaskedLM) (model EuroBertConfig)
- **flaubert** -- [FlaubertWithLMHeadModel](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertWithLMHeadModel) (model FlaubertConfig)
- **fnet** -- [FNetForMaskedLM](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForMaskedLM) (model FNetConfig)
- **funnel** -- [FunnelForMaskedLM](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForMaskedLM) (model FunnelConfig)
- **gte** -- [GteForMaskedLM](/docs/transformers/main/id/model_doc/gte#transformers.GteForMaskedLM) (model GteConfig)
- **ibert** -- [IBertForMaskedLM](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMaskedLM) (model IBertConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3ForMaskedLM](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForMaskedLM) (model JinaEmbeddingsV3Config)
- **layoutlm** -- [LayoutLMForMaskedLM](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForMaskedLM) (model LayoutLMConfig)
- **longformer** -- [LongformerForMaskedLM](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMaskedLM) (model LongformerConfig)
- **luke** -- [LukeForMaskedLM](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMaskedLM) (model LukeConfig)
- **mbart** -- [MBartForConditionalGeneration](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForConditionalGeneration) (model MBartConfig)
- **megatron-bert** -- [MegatronBertForMaskedLM](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForMaskedLM) (model MegatronBertConfig)
- **mobilebert** -- [MobileBertForMaskedLM](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForMaskedLM) (model MobileBertConfig)
- **modernbert** -- [ModernBertForMaskedLM](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForMaskedLM) (model ModernBertConfig)
- **modernvbert** -- [ModernVBertForMaskedLM](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForMaskedLM) (model ModernVBertConfig)
- **mpnet** -- [MPNetForMaskedLM](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMaskedLM) (model MPNetConfig)
- **mra** -- [MraForMaskedLM](/docs/transformers/main/id/model_doc/mra#transformers.MraForMaskedLM) (model MraConfig)
- **mvp** -- [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (model MvpConfig)
- **neomme** -- [NeoMMEForMaskedLM](/docs/transformers/main/id/model_doc/neomme#transformers.NeoMMEForMaskedLM) (model NeoMMEConfig)
- **nomic_bert** -- [NomicBertForMaskedLM](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForMaskedLM) (model NomicBertConfig)
- **nystromformer** -- [NystromformerForMaskedLM](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForMaskedLM) (model NystromformerConfig)
- **perceiver** -- [PerceiverForMaskedLM](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForMaskedLM) (model PerceiverConfig)
- **reformer** -- [ReformerForMaskedLM](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForMaskedLM) (model ReformerConfig)
- **rembert** -- [RemBertForMaskedLM](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForMaskedLM) (model RemBertConfig)
- **roberta** -- [RobertaForMaskedLM](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMaskedLM) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForMaskedLM](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMaskedLM) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForMaskedLM](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForMaskedLM) (model RoCBertConfig)
- **roformer** -- [RoFormerForMaskedLM](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForMaskedLM) (model RoFormerConfig)
- **squeezebert** -- [SqueezeBertForMaskedLM](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMaskedLM) (model SqueezeBertConfig)
- **tapas** -- [TapasForMaskedLM](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForMaskedLM) (model TapasConfig)
- **ultrabert** -- [UltrabertForMaskedLM](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForMaskedLM) (model UltrabertConfig)
- **ultravbert** -- [UltraVBertForMaskedLM](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForMaskedLM) (model UltraVBertConfig)
- **xlm** -- [XLMWithLMHeadModel](/docs/transformers/main/id/model_doc/xlm#transformers.XLMWithLMHeadModel) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMaskedLM) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForMaskedLM](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMaskedLM) (model XLMRobertaXLConfig)
- **xmod** -- [XmodForMaskedLM](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMaskedLM) (model XmodConfig)
- **yoso** -- [YosoForMaskedLM](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForMaskedLM) (model YosoConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMaskedLM

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForMaskedLM.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForMaskedLM.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForMaskGeneration[[transformers.AutoModelForMaskGeneration]]

#### transformers.AutoModelForMaskGeneration[[transformers.AutoModelForMaskGeneration]]

```python
transformers.AutoModelForMaskGeneration(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2277)

### AutoModelForSeq2SeqLM[[transformers.AutoModelForSeq2SeqLM]]

#### transformers.AutoModelForSeq2SeqLM[[transformers.AutoModelForSeq2SeqLM]]

```python
transformers.AutoModelForSeq2SeqLM(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2335)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berurutan-ke-urutan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForSeq2SeqLM.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [AudioFlamingo3Config](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3Config): [AudioFlamingo3ForConditionalGeneration](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3ForConditionalGeneration) (model AudioFlamingo3Config) - Kelas konfigurasi [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig): [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (model BartConfig) - Kelas konfigurasi [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig): [BigBirdPegasusForConditionalGeneration](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForConditionalGeneration) (model BigBirdPegasusConfig) - Kelas konfigurasi [BlenderbotConfig](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotConfig): [BlenderbotForConditionalGeneration](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotForConditionalGeneration) (model BlenderbotConfig) - Kelas konfigurasi [BlenderbotSmallConfig](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallConfig): [BlenderbotSmallForConditionalGeneration](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallForConditionalGeneration) (model BlenderbotSmallConfig) - Kelas konfigurasi [EncoderDecoderConfig](/docs/transformers/main/id/model_doc/encoder-decoder#transformers.EncoderDecoderConfig): [EncoderDecoderModel](/docs/transformers/main/id/model_doc/encoder-decoder#transformers.EncoderDecoderModel) (model EncoderDecoderConfig) - Kelas konfigurasi [FSMTConfig](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTConfig): [FSMTForConditionalGeneration](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTForConditionalGeneration) (model FSMTConfig) - Kelas konfigurasi [GlmAsrConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrConfig): [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (model GlmAsrConfig) - Kelas konfigurasi [GraniteSpeechConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechConfig): [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (model GraniteSpeechConfig) - Kelas konfigurasi [GraniteSpeechPlusConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusConfig): [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (model GraniteSpeechPlusConfig) - Kelas konfigurasi [LEDConfig](/docs/transformers/main/id/model_doc/led#transformers.LEDConfig): [LEDForConditionalGeneration](/docs/transformers/main/id/model_doc/led#transformers.LEDForConditionalGeneration) (model LEDConfig) - Kelas konfigurasi [LongT5Config](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5Config): [LongT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5ForConditionalGeneration) (model LongT5Config) - Kelas konfigurasi [M2M100Config](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100Config): [M2M100ForConditionalGeneration](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100ForConditionalGeneration) (model M2M100Config) - Kelas konfigurasi [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig): [MBartForConditionalGeneration](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForConditionalGeneration) (model MBartConfig) - Kelas konfigurasi [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config): [MT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForConditionalGeneration) (model MT5Config) - Kelas konfigurasi [MarianConfig](/docs/transformers/main/id/model_doc/marian#transformers.MarianConfig): [MarianMTModel](/docs/transformers/main/id/model_doc/marian#transformers.MarianMTModel) (model MarianConfig) - Kelas konfigurasi [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig): [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (model MossTranscribeDiarizeConfig) - Kelas konfigurasi [MusicFlamingoConfig](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoConfig): [MusicFlamingoForConditionalGeneration](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoForConditionalGeneration) (model MusicFlamingoConfig) - Kelas konfigurasi [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig): [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (model MvpConfig) - Kelas konfigurasi [Nemotron3_5TranscribeConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeConfig): [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (model Nemotron3_5TranscribeConfig) - Kelas konfigurasi [NllbMoeConfig](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeConfig): [NllbMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeForConditionalGeneration) (model NllbMoeConfig) - Kelas konfigurasi [PLBartConfig](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartConfig): [PLBartForConditionalGeneration](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForConditionalGeneration) (model PLBartConfig) - Kelas konfigurasi [PegasusConfig](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusConfig): [PegasusForConditionalGeneration](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusForConditionalGeneration) (model PegasusConfig) - Kelas konfigurasi [PegasusXConfig](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXConfig): [PegasusXForConditionalGeneration](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXForConditionalGeneration) (model PegasusXConfig) - Kelas konfigurasi [ProphetNetConfig](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetConfig): [ProphetNetForConditionalGeneration](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetForConditionalGeneration) (model ProphetNetConfig) - Kelas konfigurasi [Qwen2AudioConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioConfig): [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (model Qwen2AudioConfig) - Kelas konfigurasi [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig): [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig) - Kelas konfigurasi [SeamlessM4TConfig](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TConfig): [SeamlessM4TForTextToText](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TForTextToText) (model SeamlessM4TConfig) - Kelas konfigurasi [SeamlessM4Tv2Config](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Config): [SeamlessM4Tv2ForTextToText](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2ForTextToText) (model SeamlessM4Tv2Config) - Kelas konfigurasi [SwitchTransformersConfig](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersConfig): [SwitchTransformersForConditionalGeneration](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersForConditionalGeneration) (model SwitchTransformersConfig) - Kelas konfigurasi [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config): [T5ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5#transformers.T5ForConditionalGeneration) (model T5Config) - Kelas konfigurasi [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config): [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (model T5Gemma2Config) - Kelas konfigurasi [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig): [T5GemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForConditionalGeneration) (model T5GemmaConfig) - Kelas konfigurasi [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config): [UMT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForConditionalGeneration) (model UMT5Config) - Kelas konfigurasi [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig): [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig) - Kelas konfigurasi [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig): [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig) - Kelas konfigurasi [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig): [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berurutan-ke-urutan) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSeq2SeqLM

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-t5/t5-base")
>>> model = AutoModelForSeq2SeqLM.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForSeq2SeqLM.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan bahasa berurutan-ke-berurutan) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **audioflamingo3** -- [AudioFlamingo3ForConditionalGeneration](/docs/transformers/main/id/model_doc/audioflamingo3#transformers.AudioFlamingo3ForConditionalGeneration) (model AudioFlamingo3Config)
- **bart** -- [BartForConditionalGeneration](/docs/transformers/main/id/model_doc/bart#transformers.BartForConditionalGeneration) (model BartConfig)
- **bigbird_pegasus** -- [BigBirdPegasusForConditionalGeneration](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForConditionalGeneration) (model BigBirdPegasusConfig)
- **blenderbot** -- [BlenderbotForConditionalGeneration](/docs/transformers/main/id/model_doc/blenderbot#transformers.BlenderbotForConditionalGeneration) (model BlenderbotConfig)
- **blenderbot-small** -- [BlenderbotSmallForConditionalGeneration](/docs/transformers/main/id/model_doc/blenderbot-small#transformers.BlenderbotSmallForConditionalGeneration) (model BlenderbotSmallConfig)
- **encoder-decoder** -- [EncoderDecoderModel](/docs/transformers/main/id/model_doc/encoder-decoder#transformers.EncoderDecoderModel) (model EncoderDecoderConfig)
- **fsmt** -- [FSMTForConditionalGeneration](/docs/transformers/main/id/model_doc/fsmt#transformers.FSMTForConditionalGeneration) (model FSMTConfig)
- **glmasr** -- [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (model GlmAsrConfig)
- **granite_speech** -- [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (model GraniteSpeechConfig)
- **granite_speech_plus** -- [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (model GraniteSpeechPlusConfig)
- **led** -- [LEDForConditionalGeneration](/docs/transformers/main/id/model_doc/led#transformers.LEDForConditionalGeneration) (model LEDConfig)
- **longt5** -- [LongT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/longt5#transformers.LongT5ForConditionalGeneration) (model LongT5Config)
- **m2m_100** -- [M2M100ForConditionalGeneration](/docs/transformers/main/id/model_doc/m2m_100#transformers.M2M100ForConditionalGeneration) (model M2M100Config)
- **marian** -- [MarianMTModel](/docs/transformers/main/id/model_doc/marian#transformers.MarianMTModel) (model MarianConfig)
- **mbart** -- [MBartForConditionalGeneration](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForConditionalGeneration) (model MBartConfig)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (model MossTranscribeDiarizeConfig)
- **mt5** -- [MT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForConditionalGeneration) (model MT5Config)
- **musicflamingo** -- [MusicFlamingoForConditionalGeneration](/docs/transformers/main/id/model_doc/musicflamingo#transformers.MusicFlamingoForConditionalGeneration) (model MusicFlamingoConfig)
- **mvp** -- [MvpForConditionalGeneration](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForConditionalGeneration) (model MvpConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (model Nemotron3_5TranscribeConfig)
- **nllb-moe** -- [NllbMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/nllb-moe#transformers.NllbMoeForConditionalGeneration) (model NllbMoeConfig)
- **pegasus** -- [PegasusForConditionalGeneration](/docs/transformers/main/id/model_doc/pegasus#transformers.PegasusForConditionalGeneration) (model PegasusConfig)
- **pegasus_x** -- [PegasusXForConditionalGeneration](/docs/transformers/main/id/model_doc/pegasus_x#transformers.PegasusXForConditionalGeneration) (model PegasusXConfig)
- **plbart** -- [PLBartForConditionalGeneration](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForConditionalGeneration) (model PLBartConfig)
- **prophetnet** -- [ProphetNetForConditionalGeneration](/docs/transformers/main/id/model_doc/prophetnet#transformers.ProphetNetForConditionalGeneration) (model ProphetNetConfig)
- **qwen2_audio** -- [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (model Qwen2AudioConfig)
- **qwen3_asr** -- [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig)
- **seamless_m4t** -- [SeamlessM4TForTextToText](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TForTextToText) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4Tv2ForTextToText](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2ForTextToText) (model SeamlessM4Tv2Config)
- **switch_transformers** -- [SwitchTransformersForConditionalGeneration](/docs/transformers/main/id/model_doc/switch_transformers#transformers.SwitchTransformersForConditionalGeneration) (model SwitchTransformersConfig)
- **t5** -- [T5ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5#transformers.T5ForConditionalGeneration) (model T5Config)
- **t5gemma** -- [T5GemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForConditionalGeneration) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (model T5Gemma2Config)
- **umt5** -- [UMT5ForConditionalGeneration](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForConditionalGeneration) (model UMT5Config)
- **vibevoice_asr** -- [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig)
- **voxtral** -- [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSeq2SeqLM

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForSeq2SeqLM.from_pretrained("google-t5/t5-base")

>>> # Update configuration during loading
>>> model = AutoModelForSeq2SeqLM.from_pretrained("google-t5/t5-base", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForSequenceClassification[[transformers.AutoModelForSequenceClassification]]

#### transformers.AutoModelForSequenceClassification[[transformers.AutoModelForSequenceClassification]]

```python
transformers.AutoModelForSequenceClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2346)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi urutan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForSequenceClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AXK1Config](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Config) configuration class: [AXK1ForSequenceClassification](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForSequenceClassification) (AXK1Config model) - [AXK2Config](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Config) configuration class: [AXK2ForSequenceClassification](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForSequenceClassification) (AXK2Config model) - [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig) configuration class: [AlbertForSequenceClassification](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForSequenceClassification) (AlbertConfig model) - [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig) configuration class: [ArceeForSequenceClassification](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForSequenceClassification) (ArceeConfig model) - [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig) configuration class: [BartForSequenceClassification](/docs/transformers/main/id/model_doc/bart#transformers.BartForSequenceClassification) (BartConfig model) - [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) configuration class: [BertForSequenceClassification](/docs/transformers/main/id/model_doc/bert#transformers.BertForSequenceClassification) (BertConfig model) - [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) configuration class: [BigBirdForSequenceClassification](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForSequenceClassification) (BigBirdConfig model) - [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig) configuration class: [BigBirdPegasusForSequenceClassification](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForSequenceClassification) (BigBirdPegasusConfig model) - [BioGptConfig](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptConfig) configuration class: [BioGptForSequenceClassification](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForSequenceClassification) (BioGptConfig model) - [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) configuration class: [BloomForSequenceClassification](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForSequenceClassification) (BloomConfig model) - [CTRLConfig](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLConfig) configuration class: [CTRLForSequenceClassification](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLForSequenceClassification) (CTRLConfig model) - [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) configuration class: [CamembertForSequenceClassification](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForSequenceClassification) (CamembertConfig model) - [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig) configuration class: [CanineForSequenceClassification](/docs/transformers/main/id/model_doc/canine#transformers.CanineForSequenceClassification) (CanineConfig model) - [CohereCompassTextConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextConfig) configuration class: [CohereCompassTextForSequenceClassification](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextForSequenceClassification) (CohereCompassTextConfig model) - [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig) configuration class: [ConvBertForSequenceClassification](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForSequenceClassification) (ConvBertConfig model) - [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) configuration class: [Data2VecTextForSequenceClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForSequenceClassification) (Data2VecTextConfig model) - [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig) configuration class: [DebertaForSequenceClassification](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForSequenceClassification) (DebertaConfig model) - [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config) configuration class: [DebertaV2ForSequenceClassification](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForSequenceClassification) (DebertaV2Config model) - [DeepseekV2Config](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2Config) configuration class: [DeepseekV2ForSequenceClassification](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2ForSequenceClassification) (DeepseekV2Config model) - [DeepseekV3Config](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Config) configuration class: [DeepseekV3ForSequenceClassification](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForSequenceClassification) (DeepseekV3Config model) - [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig) configuration class: [DiffLlamaForSequenceClassification](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForSequenceClassification) (DiffLlamaConfig model) - [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig) configuration class: [DistilBertForSequenceClassification](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForSequenceClassification) (DistilBertConfig model) - [DogeConfig](/docs/transformers/main/id/model_doc/doge#transformers.DogeConfig) configuration class: [DogeForSequenceClassification](/docs/transformers/main/id/model_doc/doge#transformers.DogeForSequenceClassification) (DogeConfig model) - [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) configuration class: [ElectraForSequenceClassification](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForSequenceClassification) (ElectraConfig model) - [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) configuration class: [ErnieForSequenceClassification](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForSequenceClassification) (ErnieConfig model) - [EsmConfig](/docs/transformers/main/id/model_doc/esm#transformers.EsmConfig) configuration class: [EsmForSequenceClassification](/docs/transformers/main/id/model_doc/esm#transformers.EsmForSequenceClassification) (EsmConfig model) - [EsmcConfig](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcConfig) configuration class: [EsmcForSequenceClassification](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForSequenceClassification) (EsmcConfig model) - [EuroBertConfig](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertConfig) configuration class: [EuroBertForSequenceClassification](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForSequenceClassification) (EuroBertConfig model) - [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) configuration class: [Exaone4ForSequenceClassification](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForSequenceClassification) (Exaone4Config model) - [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig) configuration class: [FNetForSequenceClassification](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForSequenceClassification) (FNetConfig model) - [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig) configuration class: [FalconForSequenceClassification](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForSequenceClassification) (FalconConfig model) - [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig) configuration class: [FlaubertForSequenceClassification](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForSequenceClassification) (FlaubertConfig model) - [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig) configuration class: [FunnelForSequenceClassification](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForSequenceClassification) (FunnelConfig model) - [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) configuration class: [GPT2ForSequenceClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForSequenceClassification) (GPT2Config model) - [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) configuration class: [GPTBigCodeForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForSequenceClassification) (GPTBigCodeConfig model) - [GPTJConfig](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJConfig) configuration class: [GPTJForSequenceClassification](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForSequenceClassification) (GPTJConfig model) - [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig) configuration class: [GPTNeoForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForSequenceClassification) (GPTNeoConfig model) - [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig) configuration class: [GPTNeoXForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForSequenceClassification) (GPTNeoXConfig model) - [Gemma2Config](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Config) configuration class: [Gemma2ForSequenceClassification](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForSequenceClassification) (Gemma2Config model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3ForSequenceClassification](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForSequenceClassification) (Gemma3Config model) - [Gemma3TextConfig](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextConfig) configuration class: [Gemma3TextForSequenceClassification](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextForSequenceClassification) (Gemma3TextConfig model) - [GemmaConfig](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaConfig) configuration class: [GemmaForSequenceClassification](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForSequenceClassification) (GemmaConfig model) - [Glm4Config](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Config) configuration class: [Glm4ForSequenceClassification](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForSequenceClassification) (Glm4Config model) - [GlmConfig](/docs/transformers/main/id/model_doc/glm#transformers.GlmConfig) configuration class: [GlmForSequenceClassification](/docs/transformers/main/id/model_doc/glm#transformers.GlmForSequenceClassification) (GlmConfig model) - [GptOssConfig](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssConfig) configuration class: [GptOssForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForSequenceClassification) (GptOssConfig model) - [GteConfig](/docs/transformers/main/id/model_doc/gte#transformers.GteConfig) configuration class: [GteForSequenceClassification](/docs/transformers/main/id/model_doc/gte#transformers.GteForSequenceClassification) (GteConfig model) - [HeliumConfig](/docs/transformers/main/id/model_doc/helium#transformers.HeliumConfig) configuration class: [HeliumForSequenceClassification](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForSequenceClassification) (HeliumConfig model) - [HunYuanDenseV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1Config) configuration class: [HunYuanDenseV1ForSequenceClassification](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1ForSequenceClassification) (HunYuanDenseV1Config model) - [HunYuanMoEV1Config](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1Config) configuration class: [HunYuanMoEV1ForSequenceClassification](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1ForSequenceClassification) (HunYuanMoEV1Config model) - [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig) configuration class: [IBertForSequenceClassification](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForSequenceClassification) (IBertConfig model) - [JambaConfig](/docs/transformers/main/id/model_doc/jamba#transformers.JambaConfig) configuration class: [JambaForSequenceClassification](/docs/transformers/main/id/model_doc/jamba#transformers.JambaForSequenceClassification) (JambaConfig model) - [JetMoeConfig](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeConfig) configuration class: [JetMoeForSequenceClassification](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeForSequenceClassification) (JetMoeConfig model) - [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config) configuration class: [JinaEmbeddingsV3ForSequenceClassification](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForSequenceClassification) (JinaEmbeddingsV3Config model) - [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig) configuration class: [LayoutLMForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForSequenceClassification) (LayoutLMConfig model) - [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config) configuration class: [LayoutLMv2ForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForSequenceClassification) (LayoutLMv2Config model) - [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config) configuration class: [LayoutLMv3ForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForSequenceClassification) (LayoutLMv3Config model) - [LiltConfig](/docs/transformers/main/id/model_doc/lilt#transformers.LiltConfig) configuration class: [LiltForSequenceClassification](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForSequenceClassification) (LiltConfig model) - [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig) configuration class: [LlamaForSequenceClassification](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaForSequenceClassification) (LlamaConfig model) - [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig) configuration class: [LongformerForSequenceClassification](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForSequenceClassification) (LongformerConfig model) - [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig) configuration class: [LukeForSequenceClassification](/docs/transformers/main/id/model_doc/luke#transformers.LukeForSequenceClassification) (LukeConfig model) - [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig) configuration class: [MBartForSequenceClassification](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForSequenceClassification) (MBartConfig model) - [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig) configuration class: [MPNetForSequenceClassification](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForSequenceClassification) (MPNetConfig model) - [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config) configuration class: [MT5ForSequenceClassification](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForSequenceClassification) (MT5Config model) - [MarkupLMConfig](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMConfig) configuration class: [MarkupLMForSequenceClassification](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForSequenceClassification) (MarkupLMConfig model) - [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) configuration class: [MegatronBertForSequenceClassification](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForSequenceClassification) (MegatronBertConfig model) - [MiniCPM3Config](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3Config) configuration class: [MiniCPM3ForSequenceClassification](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3ForSequenceClassification) (MiniCPM3Config model) - [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig) configuration class: [MiniMaxForSequenceClassification](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForSequenceClassification) (MiniMaxConfig model) - [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config) configuration class: [Ministral3ForSequenceClassification](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForSequenceClassification) (Ministral3Config model) - [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig) configuration class: [MinistralForSequenceClassification](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForSequenceClassification) (MinistralConfig model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4ForSequenceClassification](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForSequenceClassification) (Mistral4Config model) - [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig) configuration class: [MistralForSequenceClassification](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForSequenceClassification) (MistralConfig model) - [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig) configuration class: [MixtralForSequenceClassification](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForSequenceClassification) (MixtralConfig model) - [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig) configuration class: [MobileBertForSequenceClassification](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForSequenceClassification) (MobileBertConfig model) - [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig) configuration class: [ModernBertForSequenceClassification](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForSequenceClassification) (ModernBertConfig model) - [ModernBertDecoderConfig](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderConfig) configuration class: [ModernBertDecoderForSequenceClassification](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderForSequenceClassification) (ModernBertDecoderConfig model) - [ModernVBertConfig](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertConfig) configuration class: [ModernVBertForSequenceClassification](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForSequenceClassification) (ModernVBertConfig model) - [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) configuration class: [MptForSequenceClassification](/docs/transformers/main/id/model_doc/mpt#transformers.MptForSequenceClassification) (MptConfig model) - [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig) configuration class: [MraForSequenceClassification](/docs/transformers/main/id/model_doc/mra#transformers.MraForSequenceClassification) (MraConfig model) - [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig) configuration class: [MvpForSequenceClassification](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForSequenceClassification) (MvpConfig model) - [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig) configuration class: [NemotronForSequenceClassification](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForSequenceClassification) (NemotronConfig model) - [NomicBertConfig](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertConfig) configuration class: [NomicBertForSequenceClassification](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForSequenceClassification) (NomicBertConfig model) - [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig) configuration class: [NystromformerForSequenceClassification](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForSequenceClassification) (NystromformerConfig model) - [OPTConfig](/docs/transformers/main/id/model_doc/opt#transformers.OPTConfig) configuration class: [OPTForSequenceClassification](/docs/transformers/main/id/model_doc/opt#transformers.OPTForSequenceClassification) (OPTConfig model) - [Olmo2Config](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2Config) configuration class: [Olmo2ForSequenceClassification](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2ForSequenceClassification) (Olmo2Config model) - [Olmo3Config](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3Config) configuration class: [Olmo3ForSequenceClassification](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3ForSequenceClassification) (Olmo3Config model) - [OlmoConfig](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoConfig) configuration class: [OlmoForSequenceClassification](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoForSequenceClassification) (OlmoConfig model) - [OpenAIGPTConfig](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTConfig) configuration class: [OpenAIGPTForSequenceClassification](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTForSequenceClassification) (OpenAIGPTConfig model) - [PLBartConfig](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartConfig) configuration class: [PLBartForSequenceClassification](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForSequenceClassification) (PLBartConfig model) - [PerceiverConfig](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverConfig) configuration class: [PerceiverForSequenceClassification](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForSequenceClassification) (PerceiverConfig model) - [PersimmonConfig](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonConfig) configuration class: [PersimmonForSequenceClassification](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForSequenceClassification) (PersimmonConfig model) - [Phi3Config](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Config) configuration class: [Phi3ForSequenceClassification](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForSequenceClassification) (Phi3Config model) - [PhiConfig](/docs/transformers/main/id/model_doc/phi#transformers.PhiConfig) configuration class: [PhiForSequenceClassification](/docs/transformers/main/id/model_doc/phi#transformers.PhiForSequenceClassification) (PhiConfig model) - [PhimoeConfig](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeConfig) configuration class: [PhimoeForSequenceClassification](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeForSequenceClassification) (PhimoeConfig model) - [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config) configuration class: [Qwen2ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForSequenceClassification) (Qwen2Config model) - [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig) configuration class: [Qwen2MoeForSequenceClassification](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForSequenceClassification) (Qwen2MoeConfig model) - [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config) configuration class: [Qwen3ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForSequenceClassification) (Qwen3Config model) - [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig) configuration class: [Qwen3MoeForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForSequenceClassification) (Qwen3MoeConfig model) - [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig) configuration class: [Qwen3NextForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForSequenceClassification) (Qwen3NextConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForSequenceClassification) (Qwen3_5Config model) - [Qwen3_5TextConfig](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextConfig) configuration class: [Qwen3_5TextForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextForSequenceClassification) (Qwen3_5TextConfig model) - [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig) configuration class: [ReformerForSequenceClassification](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForSequenceClassification) (ReformerConfig model) - [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig) configuration class: [RemBertForSequenceClassification](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForSequenceClassification) (RemBertConfig model) - [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) configuration class: [RoCBertForSequenceClassification](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForSequenceClassification) (RoCBertConfig model) - [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig) configuration class: [RoFormerForSequenceClassification](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForSequenceClassification) (RoFormerConfig model) - [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) configuration class: [RobertaForSequenceClassification](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForSequenceClassification) (RobertaConfig model) - [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) configuration class: [RobertaPreLayerNormForSequenceClassification](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForSequenceClassification) (RobertaPreLayerNormConfig model) - [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig) configuration class: [SeedOssForSequenceClassification](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForSequenceClassification) (SeedOssConfig model) - [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config) configuration class: [SmolLM3ForSequenceClassification](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForSequenceClassification) (SmolLM3Config model) - [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig) configuration class: [SqueezeBertForSequenceClassification](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForSequenceClassification) (SqueezeBertConfig model) - [StableLmConfig](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmConfig) configuration class: [StableLmForSequenceClassification](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForSequenceClassification) (StableLmConfig model) - [Starcoder2Config](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Config) configuration class: [Starcoder2ForSequenceClassification](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForSequenceClassification) (Starcoder2Config model) - [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config) configuration class: [T5ForSequenceClassification](/docs/transformers/main/id/model_doc/t5#transformers.T5ForSequenceClassification) (T5Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2ForSequenceClassification](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForSequenceClassification) (T5Gemma2Config model) - [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig) configuration class: [T5GemmaForSequenceClassification](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForSequenceClassification) (T5GemmaConfig model) - [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig) configuration class: [TapasForSequenceClassification](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForSequenceClassification) (TapasConfig model) - [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config) configuration class: [UMT5ForSequenceClassification](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForSequenceClassification) (UMT5Config model) - [UltraVBertConfig](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertConfig) configuration class: [UltraVBertForSequenceClassification](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForSequenceClassification) (UltraVBertConfig model) - [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig) configuration class: [UltrabertForSequenceClassification](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForSequenceClassification) (UltrabertConfig model) - [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) configuration class: [XLMForSequenceClassification](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForSequenceClassification) (XLMConfig model) - [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) configuration class: [XLMRobertaForSequenceClassification](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForSequenceClassification) (XLMRobertaConfig model) - [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) configuration class: [XLMRobertaXLForSequenceClassification](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForSequenceClassification) (XLMRobertaXLConfig model) - [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) configuration class: [XLNetForSequenceClassification](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForSequenceClassification) (XLNetConfig model) - [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) configuration class: [XmodForSequenceClassification](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForSequenceClassification) (XmodConfig model) - [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig) configuration class: [YosoForSequenceClassification](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForSequenceClassification) (YosoConfig model) - [Zamba2Config](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2Config) configuration class: [Zamba2ForSequenceClassification](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2ForSequenceClassification) (Zamba2Config model) - [ZambaConfig](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaConfig) configuration class: [ZambaForSequenceClassification](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaForSequenceClassification) (ZambaConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi urutan) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSequenceClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForSequenceClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForSequenceClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi urutan) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForSequenceClassification](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForSequenceClassification) (model AlbertConfig)
- **arcee** -- [ArceeForSequenceClassification](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForSequenceClassification) (model ArceeConfig)
- **axk1** -- [AXK1ForSequenceClassification](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForSequenceClassification) (model AXK1Config)
- **axk2** -- [AXK2ForSequenceClassification](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForSequenceClassification) (model AXK2Config)
- **bart** -- [BartForSequenceClassification](/docs/transformers/main/id/model_doc/bart#transformers.BartForSequenceClassification) (model BartConfig)
- **bert** -- [BertForSequenceClassification](/docs/transformers/main/id/model_doc/bert#transformers.BertForSequenceClassification) (model BertConfig)
- **big_bird** -- [BigBirdForSequenceClassification](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForSequenceClassification) (model BigBirdConfig)
- **bigbird_pegasus** -- [BigBirdPegasusForSequenceClassification](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForSequenceClassification) (model BigBirdPegasusConfig)
- **biogpt** -- [BioGptForSequenceClassification](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForSequenceClassification) (model BioGptConfig)
- **bloom** -- [BloomForSequenceClassification](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForSequenceClassification) (model BloomConfig)
- **camembert** -- [CamembertForSequenceClassification](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForSequenceClassification) (model CamembertConfig)
- **canine** -- [CanineForSequenceClassification](/docs/transformers/main/id/model_doc/canine#transformers.CanineForSequenceClassification) (model CanineConfig)
- **cohere_compass_text** -- [CohereCompassTextForSequenceClassification](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassTextForSequenceClassification) (model CohereCompassTextConfig)
- **convbert** -- [ConvBertForSequenceClassification](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForSequenceClassification) (model ConvBertConfig)
- **ctrl** -- [CTRLForSequenceClassification](/docs/transformers/main/id/model_doc/ctrl#transformers.CTRLForSequenceClassification) (model CTRLConfig)
- **data2vec-text** -- [Data2VecTextForSequenceClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForSequenceClassification) (model Data2VecTextConfig)
- **deberta** -- [DebertaForSequenceClassification](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForSequenceClassification) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2ForSequenceClassification](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForSequenceClassification) (model DebertaV2Config)
- **deepseek_v2** -- [DeepseekV2ForSequenceClassification](/docs/transformers/main/id/model_doc/deepseek_v2#transformers.DeepseekV2ForSequenceClassification) (model DeepseekV2Config)
- **deepseek_v3** -- [DeepseekV3ForSequenceClassification](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForSequenceClassification) (model DeepseekV3Config)
- **diffllama** -- [DiffLlamaForSequenceClassification](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForSequenceClassification) (model DiffLlamaConfig)
- **distilbert** -- [DistilBertForSequenceClassification](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForSequenceClassification) (model DistilBertConfig)
- **doge** -- [DogeForSequenceClassification](/docs/transformers/main/id/model_doc/doge#transformers.DogeForSequenceClassification) (model DogeConfig)
- **electra** -- [ElectraForSequenceClassification](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForSequenceClassification) (model ElectraConfig)
- **ernie** -- [ErnieForSequenceClassification](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForSequenceClassification) (model ErnieConfig)
- **esm** -- [EsmForSequenceClassification](/docs/transformers/main/id/model_doc/esm#transformers.EsmForSequenceClassification) (model EsmConfig)
- **esmc** -- [EsmcForSequenceClassification](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForSequenceClassification) (model EsmcConfig)
- **eurobert** -- [EuroBertForSequenceClassification](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForSequenceClassification) (model EuroBertConfig)
- **exaone4** -- [Exaone4ForSequenceClassification](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForSequenceClassification) (model Exaone4Config)
- **falcon** -- [FalconForSequenceClassification](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForSequenceClassification) (model FalconConfig)
- **flaubert** -- [FlaubertForSequenceClassification](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForSequenceClassification) (model FlaubertConfig)
- **fnet** -- [FNetForSequenceClassification](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForSequenceClassification) (model FNetConfig)
- **funnel** -- [FunnelForSequenceClassification](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForSequenceClassification) (model FunnelConfig)
- **gemma** -- [GemmaForSequenceClassification](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForSequenceClassification) (model GemmaConfig)
- **gemma2** -- [Gemma2ForSequenceClassification](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForSequenceClassification) (model Gemma2Config)
- **gemma3** -- [Gemma3ForSequenceClassification](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForSequenceClassification) (model Gemma3Config)
- **gemma3_text** -- [Gemma3TextForSequenceClassification](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3TextForSequenceClassification) (model Gemma3TextConfig)
- **glm** -- [GlmForSequenceClassification](/docs/transformers/main/id/model_doc/glm#transformers.GlmForSequenceClassification) (model GlmConfig)
- **glm4** -- [Glm4ForSequenceClassification](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForSequenceClassification) (model Glm4Config)
- **gpt-sw3** -- [GPT2ForSequenceClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForSequenceClassification) (model GPT2Config)
- **gpt2** -- [GPT2ForSequenceClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForSequenceClassification) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForSequenceClassification) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPTNeoForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForSequenceClassification) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForSequenceClassification) (model GPTNeoXConfig)
- **gpt_oss** -- [GptOssForSequenceClassification](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForSequenceClassification) (model GptOssConfig)
- **gptj** -- [GPTJForSequenceClassification](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForSequenceClassification) (model GPTJConfig)
- **gte** -- [GteForSequenceClassification](/docs/transformers/main/id/model_doc/gte#transformers.GteForSequenceClassification) (model GteConfig)
- **helium** -- [HeliumForSequenceClassification](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForSequenceClassification) (model HeliumConfig)
- **hunyuan_v1_dense** -- [HunYuanDenseV1ForSequenceClassification](/docs/transformers/main/id/model_doc/hunyuan_v1_dense#transformers.HunYuanDenseV1ForSequenceClassification) (model HunYuanDenseV1Config)
- **hunyuan_v1_moe** -- [HunYuanMoEV1ForSequenceClassification](/docs/transformers/main/id/model_doc/hunyuan_v1_moe#transformers.HunYuanMoEV1ForSequenceClassification) (model HunYuanMoEV1Config)
- **ibert** -- [IBertForSequenceClassification](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForSequenceClassification) (model IBertConfig)
- **jamba** -- [JambaForSequenceClassification](/docs/transformers/main/id/model_doc/jamba#transformers.JambaForSequenceClassification) (model JambaConfig)
- **jetmoe** -- [JetMoeForSequenceClassification](/docs/transformers/main/id/model_doc/jetmoe#transformers.JetMoeForSequenceClassification) (model JetMoeConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3ForSequenceClassification](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForSequenceClassification) (model JinaEmbeddingsV3Config)
- **layoutlm** -- [LayoutLMForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForSequenceClassification) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2ForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForSequenceClassification) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3ForSequenceClassification](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForSequenceClassification) (model LayoutLMv3Config)
- **lilt** -- [LiltForSequenceClassification](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForSequenceClassification) (model LiltConfig)
- **llama** -- [LlamaForSequenceClassification](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaForSequenceClassification) (model LlamaConfig)
- **longformer** -- [LongformerForSequenceClassification](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForSequenceClassification) (model LongformerConfig)
- **luke** -- [LukeForSequenceClassification](/docs/transformers/main/id/model_doc/luke#transformers.LukeForSequenceClassification) (model LukeConfig)
- **markuplm** -- [MarkupLMForSequenceClassification](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForSequenceClassification) (model MarkupLMConfig)
- **mbart** -- [MBartForSequenceClassification](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForSequenceClassification) (model MBartConfig)
- **megatron-bert** -- [MegatronBertForSequenceClassification](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForSequenceClassification) (model MegatronBertConfig)
- **minicpm3** -- [MiniCPM3ForSequenceClassification](/docs/transformers/main/id/model_doc/minicpm3#transformers.MiniCPM3ForSequenceClassification) (model MiniCPM3Config)
- **minimax** -- [MiniMaxForSequenceClassification](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForSequenceClassification) (model MiniMaxConfig)
- **ministral** -- [MinistralForSequenceClassification](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForSequenceClassification) (model MinistralConfig)
- **ministral3** -- [Ministral3ForSequenceClassification](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForSequenceClassification) (model Ministral3Config)
- **mistral** -- [MistralForSequenceClassification](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForSequenceClassification) (model MistralConfig)
- **mistral4** -- [Mistral4ForSequenceClassification](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForSequenceClassification) (model Mistral4Config)
- **mixtral** -- [MixtralForSequenceClassification](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForSequenceClassification) (model MixtralConfig)
- **mobilebert** -- [MobileBertForSequenceClassification](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForSequenceClassification) (model MobileBertConfig)
- **modernbert** -- [ModernBertForSequenceClassification](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForSequenceClassification) (model ModernBertConfig)
- **modernbert-decoder** -- [ModernBertDecoderForSequenceClassification](/docs/transformers/main/id/model_doc/modernbert-decoder#transformers.ModernBertDecoderForSequenceClassification) (model ModernBertDecoderConfig)
- **modernvbert** -- [ModernVBertForSequenceClassification](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForSequenceClassification) (model ModernVBertConfig)
- **mpnet** -- [MPNetForSequenceClassification](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForSequenceClassification) (model MPNetConfig)
- **mpt** -- [MptForSequenceClassification](/docs/transformers/main/id/model_doc/mpt#transformers.MptForSequenceClassification) (model MptConfig)
- **mra** -- [MraForSequenceClassification](/docs/transformers/main/id/model_doc/mra#transformers.MraForSequenceClassification) (model MraConfig)
- **mt5** -- [MT5ForSequenceClassification](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForSequenceClassification) (model MT5Config)
- **mvp** -- [MvpForSequenceClassification](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForSequenceClassification) (model MvpConfig)
- **nemotron** -- [NemotronForSequenceClassification](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForSequenceClassification) (model NemotronConfig)
- **nomic_bert** -- [NomicBertForSequenceClassification](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForSequenceClassification) (model NomicBertConfig)
- **nystromformer** -- [NystromformerForSequenceClassification](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForSequenceClassification) (model NystromformerConfig)
- **olmo** -- [OlmoForSequenceClassification](/docs/transformers/main/id/model_doc/olmo#transformers.OlmoForSequenceClassification) (model OlmoConfig)
- **olmo2** -- [Olmo2ForSequenceClassification](/docs/transformers/main/id/model_doc/olmo2#transformers.Olmo2ForSequenceClassification) (model Olmo2Config)
- **olmo3** -- [Olmo3ForSequenceClassification](/docs/transformers/main/id/model_doc/olmo3#transformers.Olmo3ForSequenceClassification) (model Olmo3Config)
- **openai-gpt** -- [OpenAIGPTForSequenceClassification](/docs/transformers/main/id/model_doc/openai-gpt#transformers.OpenAIGPTForSequenceClassification) (model OpenAIGPTConfig)
- **opt** -- [OPTForSequenceClassification](/docs/transformers/main/id/model_doc/opt#transformers.OPTForSequenceClassification) (model OPTConfig)
- **perceiver** -- [PerceiverForSequenceClassification](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForSequenceClassification) (model PerceiverConfig)
- **persimmon** -- [PersimmonForSequenceClassification](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForSequenceClassification) (model PersimmonConfig)
- **phi** -- [PhiForSequenceClassification](/docs/transformers/main/id/model_doc/phi#transformers.PhiForSequenceClassification) (model PhiConfig)
- **phi3** -- [Phi3ForSequenceClassification](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForSequenceClassification) (model Phi3Config)
- **phimoe** -- [PhimoeForSequenceClassification](/docs/transformers/main/id/model_doc/phimoe#transformers.PhimoeForSequenceClassification) (model PhimoeConfig)
- **plbart** -- [PLBartForSequenceClassification](/docs/transformers/main/id/model_doc/plbart#transformers.PLBartForSequenceClassification) (model PLBartConfig)
- **qwen2** -- [Qwen2ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForSequenceClassification) (model Qwen2Config)
- **qwen2_moe** -- [Qwen2MoeForSequenceClassification](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForSequenceClassification) (model Qwen2MoeConfig)
- **qwen3** -- [Qwen3ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForSequenceClassification) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5ForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForSequenceClassification) (model Qwen3_5Config)
- **qwen3_5_text** -- [Qwen3_5TextForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5TextForSequenceClassification) (model Qwen3_5TextConfig)
- **qwen3_moe** -- [Qwen3MoeForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForSequenceClassification) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextForSequenceClassification](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForSequenceClassification) (model Qwen3NextConfig)
- **reformer** -- [ReformerForSequenceClassification](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForSequenceClassification) (model ReformerConfig)
- **rembert** -- [RemBertForSequenceClassification](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForSequenceClassification) (model RemBertConfig)
- **roberta** -- [RobertaForSequenceClassification](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForSequenceClassification) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForSequenceClassification](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForSequenceClassification) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForSequenceClassification](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForSequenceClassification) (model RoCBertConfig)
- **roformer** -- [RoFormerForSequenceClassification](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForSequenceClassification) (model RoFormerConfig)
- **seed_oss** -- [SeedOssForSequenceClassification](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForSequenceClassification) (model SeedOssConfig)
- **smollm3** -- [SmolLM3ForSequenceClassification](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForSequenceClassification) (model SmolLM3Config)
- **squeezebert** -- [SqueezeBertForSequenceClassification](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForSequenceClassification) (model SqueezeBertConfig)
- **stablelm** -- [StableLmForSequenceClassification](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForSequenceClassification) (model StableLmConfig)
- **starcoder2** -- [Starcoder2ForSequenceClassification](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForSequenceClassification) (model Starcoder2Config)
- **t5** -- [T5ForSequenceClassification](/docs/transformers/main/id/model_doc/t5#transformers.T5ForSequenceClassification) (model T5Config)
- **t5gemma** -- [T5GemmaForSequenceClassification](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForSequenceClassification) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2ForSequenceClassification](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForSequenceClassification) (model T5Gemma2Config)
- **tapas** -- [TapasForSequenceClassification](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForSequenceClassification) (model TapasConfig)
- **ultrabert** -- [UltrabertForSequenceClassification](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForSequenceClassification) (model UltrabertConfig)
- **ultravbert** -- [UltraVBertForSequenceClassification](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForSequenceClassification) (model UltraVBertConfig)
- **umt5** -- [UMT5ForSequenceClassification](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForSequenceClassification) (model UMT5Config)
- **xlm** -- [XLMForSequenceClassification](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForSequenceClassification) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForSequenceClassification](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForSequenceClassification) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForSequenceClassification](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForSequenceClassification) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetForSequenceClassification](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForSequenceClassification) (model XLNetConfig)
- **xmod** -- [XmodForSequenceClassification](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForSequenceClassification) (model XmodConfig)
- **yoso** -- [YosoForSequenceClassification](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForSequenceClassification) (model YosoConfig)
- **zamba** -- [ZambaForSequenceClassification](/docs/transformers/main/id/model_doc/zamba#transformers.ZambaForSequenceClassification) (model ZambaConfig)
- **zamba2** -- [Zamba2ForSequenceClassification](/docs/transformers/main/id/model_doc/zamba2#transformers.Zamba2ForSequenceClassification) (model Zamba2Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSequenceClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForMultipleChoice[[transformers.AutoModelForMultipleChoice]]

#### transformers.AutoModelForMultipleChoice[[transformers.AutoModelForMultipleChoice]]

```python
transformers.AutoModelForMultipleChoice(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2402)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pilihan ganda) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForMultipleChoice.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig): [AlbertForMultipleChoice](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForMultipleChoice) (model AlbertConfig) - Kelas konfigurasi [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig): [BertForMultipleChoice](/docs/transformers/main/id/model_doc/bert#transformers.BertForMultipleChoice) (model BertConfig) - Kelas konfigurasi [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig): [BigBirdForMultipleChoice](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForMultipleChoice) (model BigBirdConfig) - Kelas konfigurasi [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig): [CamembertForMultipleChoice](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMultipleChoice) (model CamembertConfig) - Kelas konfigurasi [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig): [CanineForMultipleChoice](/docs/transformers/main/id/model_doc/canine#transformers.CanineForMultipleChoice) (model CanineConfig) - Kelas konfigurasi [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig): [ConvBertForMultipleChoice](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForMultipleChoice) (model ConvBertConfig) - Kelas konfigurasi [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig): [Data2VecTextForMultipleChoice](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMultipleChoice) (model Data2VecTextConfig) - Kelas konfigurasi [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config): [DebertaV2ForMultipleChoice](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMultipleChoice) (model DebertaV2Config) - Kelas konfigurasi [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig): [DistilBertForMultipleChoice](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMultipleChoice) (model DistilBertConfig) - Kelas konfigurasi [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig): [ElectraForMultipleChoice](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForMultipleChoice) (model ElectraConfig) - Kelas konfigurasi [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig): [ErnieForMultipleChoice](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForMultipleChoice) (model ErnieConfig) - Kelas konfigurasi [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig): [FNetForMultipleChoice](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForMultipleChoice) (model FNetConfig) - Kelas konfigurasi [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig): [FlaubertForMultipleChoice](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForMultipleChoice) (model FlaubertConfig) - Kelas konfigurasi [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig): [FunnelForMultipleChoice](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForMultipleChoice) (model FunnelConfig) - Kelas konfigurasi [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig): [IBertForMultipleChoice](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMultipleChoice) (model IBertConfig) - Kelas konfigurasi [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig): [LongformerForMultipleChoice](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMultipleChoice) (model LongformerConfig) - Kelas konfigurasi [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig): [LukeForMultipleChoice](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMultipleChoice) (model LukeConfig) - Kelas konfigurasi [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig): [MPNetForMultipleChoice](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMultipleChoice) (model MPNetConfig) - Kelas konfigurasi [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig): [MegatronBertForMultipleChoice](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForMultipleChoice) (model MegatronBertConfig) - Kelas konfigurasi [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig): [MobileBertForMultipleChoice](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForMultipleChoice) (model MobileBertConfig) - Kelas konfigurasi [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig): [ModernBertForMultipleChoice](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForMultipleChoice) (model ModernBertConfig) - Kelas konfigurasi [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig): [MraForMultipleChoice](/docs/transformers/main/id/model_doc/mra#transformers.MraForMultipleChoice) (model MraConfig) - Kelas konfigurasi [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig): [NystromformerForMultipleChoice](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForMultipleChoice) (model NystromformerConfig) - Kelas konfigurasi [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig): [RemBertForMultipleChoice](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForMultipleChoice) (model RemBertConfig) - Kelas konfigurasi [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig): [RoCBertForMultipleChoice](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForMultipleChoice) (model RoCBertConfig) - Kelas konfigurasi [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig): [RoFormerForMultipleChoice](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForMultipleChoice) (model RoFormerConfig) - Kelas konfigurasi [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig): [RobertaForMultipleChoice](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMultipleChoice) (model RobertaConfig) - Kelas konfigurasi [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig): [RobertaPreLayerNormForMultipleChoice](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMultipleChoice) (model RobertaPreLayerNormConfig) - Kelas konfigurasi [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig): [SqueezeBertForMultipleChoice](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMultipleChoice) (model SqueezeBertConfig) - Kelas konfigurasi [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig): [XLMForMultipleChoice](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForMultipleChoice) (model XLMConfig) - Kelas konfigurasi [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig): [XLMRobertaForMultipleChoice](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMultipleChoice) (model XLMRobertaConfig) - Kelas konfigurasi [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig): [XLMRobertaXLForMultipleChoice](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMultipleChoice) (model XLMRobertaXLConfig) - Kelas konfigurasi [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig): [XLNetForMultipleChoice](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForMultipleChoice) (model XLNetConfig) - Kelas konfigurasi [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig): [XmodForMultipleChoice](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMultipleChoice) (model XmodConfig) - Kelas konfigurasi [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig): [YosoForMultipleChoice](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForMultipleChoice) (model YosoConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pilihan ganda) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMultipleChoice

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForMultipleChoice.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForMultipleChoice.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pilihan ganda) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForMultipleChoice](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForMultipleChoice) (model AlbertConfig)
- **bert** -- [BertForMultipleChoice](/docs/transformers/main/id/model_doc/bert#transformers.BertForMultipleChoice) (model BertConfig)
- **big_bird** -- [BigBirdForMultipleChoice](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForMultipleChoice) (model BigBirdConfig)
- **camembert** -- [CamembertForMultipleChoice](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForMultipleChoice) (model CamembertConfig)
- **canine** -- [CanineForMultipleChoice](/docs/transformers/main/id/model_doc/canine#transformers.CanineForMultipleChoice) (model CanineConfig)
- **convbert** -- [ConvBertForMultipleChoice](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForMultipleChoice) (model ConvBertConfig)
- **data2vec-text** -- [Data2VecTextForMultipleChoice](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForMultipleChoice) (model Data2VecTextConfig)
- **deberta-v2** -- [DebertaV2ForMultipleChoice](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForMultipleChoice) (model DebertaV2Config)
- **distilbert** -- [DistilBertForMultipleChoice](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForMultipleChoice) (model DistilBertConfig)
- **electra** -- [ElectraForMultipleChoice](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForMultipleChoice) (model ElectraConfig)
- **ernie** -- [ErnieForMultipleChoice](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForMultipleChoice) (model ErnieConfig)
- **flaubert** -- [FlaubertForMultipleChoice](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForMultipleChoice) (model FlaubertConfig)
- **fnet** -- [FNetForMultipleChoice](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForMultipleChoice) (model FNetConfig)
- **funnel** -- [FunnelForMultipleChoice](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForMultipleChoice) (model FunnelConfig)
- **ibert** -- [IBertForMultipleChoice](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForMultipleChoice) (model IBertConfig)
- **longformer** -- [LongformerForMultipleChoice](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForMultipleChoice) (model LongformerConfig)
- **luke** -- [LukeForMultipleChoice](/docs/transformers/main/id/model_doc/luke#transformers.LukeForMultipleChoice) (model LukeConfig)
- **megatron-bert** -- [MegatronBertForMultipleChoice](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForMultipleChoice) (model MegatronBertConfig)
- **mobilebert** -- [MobileBertForMultipleChoice](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForMultipleChoice) (model MobileBertConfig)
- **modernbert** -- [ModernBertForMultipleChoice](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForMultipleChoice) (model ModernBertConfig)
- **mpnet** -- [MPNetForMultipleChoice](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForMultipleChoice) (model MPNetConfig)
- **mra** -- [MraForMultipleChoice](/docs/transformers/main/id/model_doc/mra#transformers.MraForMultipleChoice) (model MraConfig)
- **nystromformer** -- [NystromformerForMultipleChoice](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForMultipleChoice) (model NystromformerConfig)
- **rembert** -- [RemBertForMultipleChoice](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForMultipleChoice) (model RemBertConfig)
- **roberta** -- [RobertaForMultipleChoice](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForMultipleChoice) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForMultipleChoice](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForMultipleChoice) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForMultipleChoice](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForMultipleChoice) (model RoCBertConfig)
- **roformer** -- [RoFormerForMultipleChoice](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForMultipleChoice) (model RoFormerConfig)
- **squeezebert** -- [SqueezeBertForMultipleChoice](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForMultipleChoice) (model SqueezeBertConfig)
- **xlm** -- [XLMForMultipleChoice](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForMultipleChoice) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForMultipleChoice](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForMultipleChoice) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForMultipleChoice](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForMultipleChoice) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetForMultipleChoice](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForMultipleChoice) (model XLNetConfig)
- **xmod** -- [XmodForMultipleChoice](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForMultipleChoice) (model XmodConfig)
- **yoso** -- [YosoForMultipleChoice](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForMultipleChoice) (model YosoConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMultipleChoice

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForMultipleChoice.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForMultipleChoice.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForNextSentencePrediction[[transformers.AutoModelForNextSentencePrediction]]

#### transformers.AutoModelForNextSentencePrediction[[transformers.AutoModelForNextSentencePrediction]]

```python
transformers.AutoModelForNextSentencePrediction(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2409)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala prediksi kalimat berikutnya) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForNextSentencePrediction.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig): [BertForNextSentencePrediction](/docs/transformers/main/id/model_doc/bert#transformers.BertForNextSentencePrediction) (model BertConfig) - Kelas konfigurasi [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig): [ErnieForNextSentencePrediction](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForNextSentencePrediction) (model ErnieConfig) - Kelas konfigurasi [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig): [FNetForNextSentencePrediction](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForNextSentencePrediction) (model FNetConfig) - Kelas konfigurasi [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig): [MegatronBertForNextSentencePrediction](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForNextSentencePrediction) (model MegatronBertConfig) - Kelas konfigurasi [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig): [MobileBertForNextSentencePrediction](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForNextSentencePrediction) (model MobileBertConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala prediksi kalimat berikutnya) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForNextSentencePrediction

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForNextSentencePrediction.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForNextSentencePrediction.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala prediksi kalimat berikutnya) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **bert** -- [BertForNextSentencePrediction](/docs/transformers/main/id/model_doc/bert#transformers.BertForNextSentencePrediction) (model BertConfig)
- **ernie** -- [ErnieForNextSentencePrediction](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForNextSentencePrediction) (model ErnieConfig)
- **fnet** -- [FNetForNextSentencePrediction](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForNextSentencePrediction) (model FNetConfig)
- **megatron-bert** -- [MegatronBertForNextSentencePrediction](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForNextSentencePrediction) (model MegatronBertConfig)
- **mobilebert** -- [MobileBertForNextSentencePrediction](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForNextSentencePrediction) (model MobileBertConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForNextSentencePrediction

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForNextSentencePrediction.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForNextSentencePrediction.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTokenClassification[[transformers.AutoModelForTokenClassification]]

#### transformers.AutoModelForTokenClassification[[transformers.AutoModelForTokenClassification]]

```python
transformers.AutoModelForTokenClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2395)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi token) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTokenClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AXK1Config](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1Config) configuration class: [AXK1ForTokenClassification](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForTokenClassification) (AXK1Config model) - [AXK2Config](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2Config) configuration class: [AXK2ForTokenClassification](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForTokenClassification) (AXK2Config model) - [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig) configuration class: [AlbertForTokenClassification](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForTokenClassification) (AlbertConfig model) - [ApertusConfig](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusConfig) configuration class: [ApertusForTokenClassification](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusForTokenClassification) (ApertusConfig model) - [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig) configuration class: [ArceeForTokenClassification](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForTokenClassification) (ArceeConfig model) - [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig) configuration class: [BertForTokenClassification](/docs/transformers/main/id/model_doc/bert#transformers.BertForTokenClassification) (BertConfig model) - [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig) configuration class: [BigBirdForTokenClassification](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForTokenClassification) (BigBirdConfig model) - [BioGptConfig](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptConfig) configuration class: [BioGptForTokenClassification](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForTokenClassification) (BioGptConfig model) - [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig) configuration class: [BloomForTokenClassification](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForTokenClassification) (BloomConfig model) - [BrosConfig](/docs/transformers/main/id/model_doc/bros#transformers.BrosConfig) configuration class: [BrosForTokenClassification](/docs/transformers/main/id/model_doc/bros#transformers.BrosForTokenClassification) (BrosConfig model) - [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig) configuration class: [CamembertForTokenClassification](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForTokenClassification) (CamembertConfig model) - [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig) configuration class: [CanineForTokenClassification](/docs/transformers/main/id/model_doc/canine#transformers.CanineForTokenClassification) (CanineConfig model) - [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig) configuration class: [ConvBertForTokenClassification](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForTokenClassification) (ConvBertConfig model) - [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig) configuration class: [Data2VecTextForTokenClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForTokenClassification) (Data2VecTextConfig model) - [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig) configuration class: [DebertaForTokenClassification](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForTokenClassification) (DebertaConfig model) - [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config) configuration class: [DebertaV2ForTokenClassification](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForTokenClassification) (DebertaV2Config model) - [DeepseekV3Config](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3Config) configuration class: [DeepseekV3ForTokenClassification](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForTokenClassification) (DeepseekV3Config model) - [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig) configuration class: [DiffLlamaForTokenClassification](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForTokenClassification) (DiffLlamaConfig model) - [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig) configuration class: [DistilBertForTokenClassification](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForTokenClassification) (DistilBertConfig model) - [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig) configuration class: [ElectraForTokenClassification](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForTokenClassification) (ElectraConfig model) - [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig) configuration class: [ErnieForTokenClassification](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForTokenClassification) (ErnieConfig model) - [EsmConfig](/docs/transformers/main/id/model_doc/esm#transformers.EsmConfig) configuration class: [EsmForTokenClassification](/docs/transformers/main/id/model_doc/esm#transformers.EsmForTokenClassification) (EsmConfig model) - [EsmcConfig](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcConfig) configuration class: [EsmcForTokenClassification](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForTokenClassification) (EsmcConfig model) - [EuroBertConfig](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertConfig) configuration class: [EuroBertForTokenClassification](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForTokenClassification) (EuroBertConfig model) - [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config) configuration class: [Exaone4ForTokenClassification](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForTokenClassification) (Exaone4Config model) - [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig) configuration class: [FNetForTokenClassification](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForTokenClassification) (FNetConfig model) - [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig) configuration class: [FalconForTokenClassification](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForTokenClassification) (FalconConfig model) - [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig) configuration class: [FlaubertForTokenClassification](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForTokenClassification) (FlaubertConfig model) - [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig) configuration class: [FunnelForTokenClassification](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForTokenClassification) (FunnelConfig model) - [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config) configuration class: [GPT2ForTokenClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForTokenClassification) (GPT2Config model) - [GPTBigCodeConfig](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeConfig) configuration class: [GPTBigCodeForTokenClassification](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForTokenClassification) (GPTBigCodeConfig model) - [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig) configuration class: [GPTNeoForTokenClassification](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForTokenClassification) (GPTNeoConfig model) - [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig) configuration class: [GPTNeoXForTokenClassification](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForTokenClassification) (GPTNeoXConfig model) - [Gemma2Config](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2Config) configuration class: [Gemma2ForTokenClassification](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForTokenClassification) (Gemma2Config model) - [GemmaConfig](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaConfig) configuration class: [GemmaForTokenClassification](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForTokenClassification) (GemmaConfig model) - [Glm4Config](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4Config) configuration class: [Glm4ForTokenClassification](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForTokenClassification) (Glm4Config model) - [GlmConfig](/docs/transformers/main/id/model_doc/glm#transformers.GlmConfig) configuration class: [GlmForTokenClassification](/docs/transformers/main/id/model_doc/glm#transformers.GlmForTokenClassification) (GlmConfig model) - [GptOssConfig](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssConfig) configuration class: [GptOssForTokenClassification](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForTokenClassification) (GptOssConfig model) - [GteConfig](/docs/transformers/main/id/model_doc/gte#transformers.GteConfig) configuration class: [GteForTokenClassification](/docs/transformers/main/id/model_doc/gte#transformers.GteForTokenClassification) (GteConfig model) - [HeliumConfig](/docs/transformers/main/id/model_doc/helium#transformers.HeliumConfig) configuration class: [HeliumForTokenClassification](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForTokenClassification) (HeliumConfig model) - [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig) configuration class: [IBertForTokenClassification](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForTokenClassification) (IBertConfig model) - [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config) configuration class: [JinaEmbeddingsV3ForTokenClassification](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForTokenClassification) (JinaEmbeddingsV3Config model) - [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig) configuration class: [LayoutLMForTokenClassification](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForTokenClassification) (LayoutLMConfig model) - [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config) configuration class: [LayoutLMv2ForTokenClassification](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForTokenClassification) (LayoutLMv2Config model) - [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config) configuration class: [LayoutLMv3ForTokenClassification](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForTokenClassification) (LayoutLMv3Config model) - [LiltConfig](/docs/transformers/main/id/model_doc/lilt#transformers.LiltConfig) configuration class: [LiltForTokenClassification](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForTokenClassification) (LiltConfig model) - [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig) configuration class: [LlamaForTokenClassification](/docs/transformers/main/id/model_doc/llama#transformers.LlamaForTokenClassification) (LlamaConfig model) - [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig) configuration class: [LongformerForTokenClassification](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForTokenClassification) (LongformerConfig model) - [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig) configuration class: [LukeForTokenClassification](/docs/transformers/main/id/model_doc/luke#transformers.LukeForTokenClassification) (LukeConfig model) - [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig) configuration class: [MPNetForTokenClassification](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForTokenClassification) (MPNetConfig model) - [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config) configuration class: [MT5ForTokenClassification](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForTokenClassification) (MT5Config model) - [MarkupLMConfig](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMConfig) configuration class: [MarkupLMForTokenClassification](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForTokenClassification) (MarkupLMConfig model) - [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig) configuration class: [MegatronBertForTokenClassification](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForTokenClassification) (MegatronBertConfig model) - [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig) configuration class: [MiniMaxForTokenClassification](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForTokenClassification) (MiniMaxConfig model) - [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config) configuration class: [Ministral3ForTokenClassification](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForTokenClassification) (Ministral3Config model) - [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig) configuration class: [MinistralForTokenClassification](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForTokenClassification) (MinistralConfig model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4ForTokenClassification](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForTokenClassification) (Mistral4Config model) - [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig) configuration class: [MistralForTokenClassification](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForTokenClassification) (MistralConfig model) - [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig) configuration class: [MixtralForTokenClassification](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForTokenClassification) (MixtralConfig model) - [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig) configuration class: [MobileBertForTokenClassification](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForTokenClassification) (MobileBertConfig model) - [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig) configuration class: [ModernBertForTokenClassification](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForTokenClassification) (ModernBertConfig model) - [ModernVBertConfig](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertConfig) configuration class: [ModernVBertForTokenClassification](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForTokenClassification) (ModernVBertConfig model) - [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig) configuration class: [MptForTokenClassification](/docs/transformers/main/id/model_doc/mpt#transformers.MptForTokenClassification) (MptConfig model) - [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig) configuration class: [MraForTokenClassification](/docs/transformers/main/id/model_doc/mra#transformers.MraForTokenClassification) (MraConfig model) - [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig) configuration class: [NemotronForTokenClassification](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForTokenClassification) (NemotronConfig model) - [NomicBertConfig](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertConfig) configuration class: [NomicBertForTokenClassification](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForTokenClassification) (NomicBertConfig model) - [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig) configuration class: [NystromformerForTokenClassification](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForTokenClassification) (NystromformerConfig model) - [OpenAIPrivacyFilterConfig](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterConfig) configuration class: [OpenAIPrivacyFilterForTokenClassification](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterForTokenClassification) (OpenAIPrivacyFilterConfig model) - [PersimmonConfig](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonConfig) configuration class: [PersimmonForTokenClassification](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForTokenClassification) (PersimmonConfig model) - [Phi3Config](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3Config) configuration class: [Phi3ForTokenClassification](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForTokenClassification) (Phi3Config model) - [PhiConfig](/docs/transformers/main/id/model_doc/phi#transformers.PhiConfig) configuration class: [PhiForTokenClassification](/docs/transformers/main/id/model_doc/phi#transformers.PhiForTokenClassification) (PhiConfig model) - [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config) configuration class: [Qwen2ForTokenClassification](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForTokenClassification) (Qwen2Config model) - [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig) configuration class: [Qwen2MoeForTokenClassification](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForTokenClassification) (Qwen2MoeConfig model) - [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig) configuration class: [Qwen3ASRForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForTokenClassification) (Qwen3ASRConfig model) - [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config) configuration class: [Qwen3ForTokenClassification](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForTokenClassification) (Qwen3Config model) - [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig) configuration class: [Qwen3MoeForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForTokenClassification) (Qwen3MoeConfig model) - [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig) configuration class: [Qwen3NextForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForTokenClassification) (Qwen3NextConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5ForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForTokenClassification) (Qwen3_5Config model) - [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig) configuration class: [RemBertForTokenClassification](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForTokenClassification) (RemBertConfig model) - [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig) configuration class: [RoCBertForTokenClassification](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForTokenClassification) (RoCBertConfig model) - [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig) configuration class: [RoFormerForTokenClassification](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForTokenClassification) (RoFormerConfig model) - [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig) configuration class: [RobertaForTokenClassification](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForTokenClassification) (RobertaConfig model) - [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig) configuration class: [RobertaPreLayerNormForTokenClassification](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForTokenClassification) (RobertaPreLayerNormConfig model) - [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig) configuration class: [SeedOssForTokenClassification](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForTokenClassification) (SeedOssConfig model) - [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config) configuration class: [SmolLM3ForTokenClassification](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForTokenClassification) (SmolLM3Config model) - [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig) configuration class: [SqueezeBertForTokenClassification](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForTokenClassification) (SqueezeBertConfig model) - [StableLmConfig](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmConfig) configuration class: [StableLmForTokenClassification](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForTokenClassification) (StableLmConfig model) - [Starcoder2Config](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2Config) configuration class: [Starcoder2ForTokenClassification](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForTokenClassification) (Starcoder2Config model) - [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config) configuration class: [T5ForTokenClassification](/docs/transformers/main/id/model_doc/t5#transformers.T5ForTokenClassification) (T5Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2ForTokenClassification](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForTokenClassification) (T5Gemma2Config model) - [T5GemmaConfig](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaConfig) configuration class: [T5GemmaForTokenClassification](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForTokenClassification) (T5GemmaConfig model) - [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config) configuration class: [UMT5ForTokenClassification](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForTokenClassification) (UMT5Config model) - [UltraVBertConfig](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertConfig) configuration class: [UltraVBertForTokenClassification](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForTokenClassification) (UltraVBertConfig model) - [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig) configuration class: [UltrabertForTokenClassification](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForTokenClassification) (UltrabertConfig model) - [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig) configuration class: [XLMForTokenClassification](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForTokenClassification) (XLMConfig model) - [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig) configuration class: [XLMRobertaForTokenClassification](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForTokenClassification) (XLMRobertaConfig model) - [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig) configuration class: [XLMRobertaXLForTokenClassification](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForTokenClassification) (XLMRobertaXLConfig model) - [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig) configuration class: [XLNetForTokenClassification](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForTokenClassification) (XLNetConfig model) - [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig) configuration class: [XmodForTokenClassification](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForTokenClassification) (XmodConfig model) - [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig) configuration class: [YosoForTokenClassification](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForTokenClassification) (YosoConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi token) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTokenClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForTokenClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTokenClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi token) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForTokenClassification](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForTokenClassification) (model AlbertConfig)
- **apertus** -- [ApertusForTokenClassification](/docs/transformers/main/id/model_doc/apertus#transformers.ApertusForTokenClassification) (model ApertusConfig)
- **arcee** -- [ArceeForTokenClassification](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForTokenClassification) (model ArceeConfig)
- **axk1** -- [AXK1ForTokenClassification](/docs/transformers/main/id/model_doc/axk1#transformers.AXK1ForTokenClassification) (model AXK1Config)
- **axk2** -- [AXK2ForTokenClassification](/docs/transformers/main/id/model_doc/axk2#transformers.AXK2ForTokenClassification) (model AXK2Config)
- **bert** -- [BertForTokenClassification](/docs/transformers/main/id/model_doc/bert#transformers.BertForTokenClassification) (model BertConfig)
- **big_bird** -- [BigBirdForTokenClassification](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForTokenClassification) (model BigBirdConfig)
- **biogpt** -- [BioGptForTokenClassification](/docs/transformers/main/id/model_doc/biogpt#transformers.BioGptForTokenClassification) (model BioGptConfig)
- **bloom** -- [BloomForTokenClassification](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForTokenClassification) (model BloomConfig)
- **bros** -- [BrosForTokenClassification](/docs/transformers/main/id/model_doc/bros#transformers.BrosForTokenClassification) (model BrosConfig)
- **camembert** -- [CamembertForTokenClassification](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForTokenClassification) (model CamembertConfig)
- **canine** -- [CanineForTokenClassification](/docs/transformers/main/id/model_doc/canine#transformers.CanineForTokenClassification) (model CanineConfig)
- **convbert** -- [ConvBertForTokenClassification](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForTokenClassification) (model ConvBertConfig)
- **data2vec-text** -- [Data2VecTextForTokenClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForTokenClassification) (model Data2VecTextConfig)
- **deberta** -- [DebertaForTokenClassification](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForTokenClassification) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2ForTokenClassification](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForTokenClassification) (model DebertaV2Config)
- **deepseek_v3** -- [DeepseekV3ForTokenClassification](/docs/transformers/main/id/model_doc/deepseek_v3#transformers.DeepseekV3ForTokenClassification) (model DeepseekV3Config)
- **diffllama** -- [DiffLlamaForTokenClassification](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForTokenClassification) (model DiffLlamaConfig)
- **distilbert** -- [DistilBertForTokenClassification](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForTokenClassification) (model DistilBertConfig)
- **electra** -- [ElectraForTokenClassification](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForTokenClassification) (model ElectraConfig)
- **ernie** -- [ErnieForTokenClassification](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForTokenClassification) (model ErnieConfig)
- **esm** -- [EsmForTokenClassification](/docs/transformers/main/id/model_doc/esm#transformers.EsmForTokenClassification) (model EsmConfig)
- **esmc** -- [EsmcForTokenClassification](/docs/transformers/main/id/model_doc/esmc#transformers.EsmcForTokenClassification) (model EsmcConfig)
- **eurobert** -- [EuroBertForTokenClassification](/docs/transformers/main/id/model_doc/eurobert#transformers.EuroBertForTokenClassification) (model EuroBertConfig)
- **exaone4** -- [Exaone4ForTokenClassification](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForTokenClassification) (model Exaone4Config)
- **falcon** -- [FalconForTokenClassification](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForTokenClassification) (model FalconConfig)
- **flaubert** -- [FlaubertForTokenClassification](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForTokenClassification) (model FlaubertConfig)
- **fnet** -- [FNetForTokenClassification](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForTokenClassification) (model FNetConfig)
- **funnel** -- [FunnelForTokenClassification](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForTokenClassification) (model FunnelConfig)
- **gemma** -- [GemmaForTokenClassification](/docs/transformers/main/id/model_doc/gemma#transformers.GemmaForTokenClassification) (model GemmaConfig)
- **gemma2** -- [Gemma2ForTokenClassification](/docs/transformers/main/id/model_doc/gemma2#transformers.Gemma2ForTokenClassification) (model Gemma2Config)
- **glm** -- [GlmForTokenClassification](/docs/transformers/main/id/model_doc/glm#transformers.GlmForTokenClassification) (model GlmConfig)
- **glm4** -- [Glm4ForTokenClassification](/docs/transformers/main/id/model_doc/glm4#transformers.Glm4ForTokenClassification) (model Glm4Config)
- **gpt-sw3** -- [GPT2ForTokenClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForTokenClassification) (model GPT2Config)
- **gpt2** -- [GPT2ForTokenClassification](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForTokenClassification) (model GPT2Config)
- **gpt_bigcode** -- [GPTBigCodeForTokenClassification](/docs/transformers/main/id/model_doc/gpt_bigcode#transformers.GPTBigCodeForTokenClassification) (model GPTBigCodeConfig)
- **gpt_neo** -- [GPTNeoForTokenClassification](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForTokenClassification) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXForTokenClassification](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForTokenClassification) (model GPTNeoXConfig)
- **gpt_oss** -- [GptOssForTokenClassification](/docs/transformers/main/id/model_doc/gpt_oss#transformers.GptOssForTokenClassification) (model GptOssConfig)
- **gte** -- [GteForTokenClassification](/docs/transformers/main/id/model_doc/gte#transformers.GteForTokenClassification) (model GteConfig)
- **helium** -- [HeliumForTokenClassification](/docs/transformers/main/id/model_doc/helium#transformers.HeliumForTokenClassification) (model HeliumConfig)
- **ibert** -- [IBertForTokenClassification](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForTokenClassification) (model IBertConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3ForTokenClassification](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForTokenClassification) (model JinaEmbeddingsV3Config)
- **layoutlm** -- [LayoutLMForTokenClassification](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForTokenClassification) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2ForTokenClassification](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForTokenClassification) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3ForTokenClassification](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForTokenClassification) (model LayoutLMv3Config)
- **lilt** -- [LiltForTokenClassification](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForTokenClassification) (model LiltConfig)
- **llama** -- [LlamaForTokenClassification](/docs/transformers/main/id/model_doc/llama#transformers.LlamaForTokenClassification) (model LlamaConfig)
- **longformer** -- [LongformerForTokenClassification](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForTokenClassification) (model LongformerConfig)
- **luke** -- [LukeForTokenClassification](/docs/transformers/main/id/model_doc/luke#transformers.LukeForTokenClassification) (model LukeConfig)
- **markuplm** -- [MarkupLMForTokenClassification](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForTokenClassification) (model MarkupLMConfig)
- **megatron-bert** -- [MegatronBertForTokenClassification](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForTokenClassification) (model MegatronBertConfig)
- **minimax** -- [MiniMaxForTokenClassification](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForTokenClassification) (model MiniMaxConfig)
- **ministral** -- [MinistralForTokenClassification](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForTokenClassification) (model MinistralConfig)
- **ministral3** -- [Ministral3ForTokenClassification](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForTokenClassification) (model Ministral3Config)
- **mistral** -- [MistralForTokenClassification](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForTokenClassification) (model MistralConfig)
- **mistral4** -- [Mistral4ForTokenClassification](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForTokenClassification) (model Mistral4Config)
- **mixtral** -- [MixtralForTokenClassification](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForTokenClassification) (model MixtralConfig)
- **mobilebert** -- [MobileBertForTokenClassification](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForTokenClassification) (model MobileBertConfig)
- **modernbert** -- [ModernBertForTokenClassification](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForTokenClassification) (model ModernBertConfig)
- **modernvbert** -- [ModernVBertForTokenClassification](/docs/transformers/main/id/model_doc/modernvbert#transformers.ModernVBertForTokenClassification) (model ModernVBertConfig)
- **mpnet** -- [MPNetForTokenClassification](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForTokenClassification) (model MPNetConfig)
- **mpt** -- [MptForTokenClassification](/docs/transformers/main/id/model_doc/mpt#transformers.MptForTokenClassification) (model MptConfig)
- **mra** -- [MraForTokenClassification](/docs/transformers/main/id/model_doc/mra#transformers.MraForTokenClassification) (model MraConfig)
- **mt5** -- [MT5ForTokenClassification](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForTokenClassification) (model MT5Config)
- **nemotron** -- [NemotronForTokenClassification](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForTokenClassification) (model NemotronConfig)
- **nomic_bert** -- [NomicBertForTokenClassification](/docs/transformers/main/id/model_doc/nomic_bert#transformers.NomicBertForTokenClassification) (model NomicBertConfig)
- **nystromformer** -- [NystromformerForTokenClassification](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForTokenClassification) (model NystromformerConfig)
- **openai_privacy_filter** -- [OpenAIPrivacyFilterForTokenClassification](/docs/transformers/main/id/model_doc/openai_privacy_filter#transformers.OpenAIPrivacyFilterForTokenClassification) (model OpenAIPrivacyFilterConfig)
- **persimmon** -- [PersimmonForTokenClassification](/docs/transformers/main/id/model_doc/persimmon#transformers.PersimmonForTokenClassification) (model PersimmonConfig)
- **phi** -- [PhiForTokenClassification](/docs/transformers/main/id/model_doc/phi#transformers.PhiForTokenClassification) (model PhiConfig)
- **phi3** -- [Phi3ForTokenClassification](/docs/transformers/main/id/model_doc/phi3#transformers.Phi3ForTokenClassification) (model Phi3Config)
- **qwen2** -- [Qwen2ForTokenClassification](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForTokenClassification) (model Qwen2Config)
- **qwen2_moe** -- [Qwen2MoeForTokenClassification](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForTokenClassification) (model Qwen2MoeConfig)
- **qwen3** -- [Qwen3ForTokenClassification](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForTokenClassification) (model Qwen3Config)
- **qwen3_5** -- [Qwen3_5ForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForTokenClassification) (model Qwen3_5Config)
- **qwen3_asr** -- [Qwen3ASRForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForTokenClassification) (model Qwen3ASRConfig)
- **qwen3_moe** -- [Qwen3MoeForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForTokenClassification) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextForTokenClassification](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForTokenClassification) (model Qwen3NextConfig)
- **rembert** -- [RemBertForTokenClassification](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForTokenClassification) (model RemBertConfig)
- **roberta** -- [RobertaForTokenClassification](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForTokenClassification) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForTokenClassification](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForTokenClassification) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForTokenClassification](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForTokenClassification) (model RoCBertConfig)
- **roformer** -- [RoFormerForTokenClassification](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForTokenClassification) (model RoFormerConfig)
- **seed_oss** -- [SeedOssForTokenClassification](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForTokenClassification) (model SeedOssConfig)
- **smollm3** -- [SmolLM3ForTokenClassification](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForTokenClassification) (model SmolLM3Config)
- **squeezebert** -- [SqueezeBertForTokenClassification](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForTokenClassification) (model SqueezeBertConfig)
- **stablelm** -- [StableLmForTokenClassification](/docs/transformers/main/id/model_doc/stablelm#transformers.StableLmForTokenClassification) (model StableLmConfig)
- **starcoder2** -- [Starcoder2ForTokenClassification](/docs/transformers/main/id/model_doc/starcoder2#transformers.Starcoder2ForTokenClassification) (model Starcoder2Config)
- **t5** -- [T5ForTokenClassification](/docs/transformers/main/id/model_doc/t5#transformers.T5ForTokenClassification) (model T5Config)
- **t5gemma** -- [T5GemmaForTokenClassification](/docs/transformers/main/id/model_doc/t5gemma#transformers.T5GemmaForTokenClassification) (model T5GemmaConfig)
- **t5gemma2** -- [T5Gemma2ForTokenClassification](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForTokenClassification) (model T5Gemma2Config)
- **ultrabert** -- [UltrabertForTokenClassification](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForTokenClassification) (model UltrabertConfig)
- **ultravbert** -- [UltraVBertForTokenClassification](/docs/transformers/main/id/model_doc/ultravbert#transformers.UltraVBertForTokenClassification) (model UltraVBertConfig)
- **umt5** -- [UMT5ForTokenClassification](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForTokenClassification) (model UMT5Config)
- **xlm** -- [XLMForTokenClassification](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForTokenClassification) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForTokenClassification](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForTokenClassification) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForTokenClassification](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForTokenClassification) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetForTokenClassification](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForTokenClassification) (model XLNetConfig)
- **xmod** -- [XmodForTokenClassification](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForTokenClassification) (model XmodConfig)
- **yoso** -- [YosoForTokenClassification](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForTokenClassification) (model YosoConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTokenClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTokenClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForTokenClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForQuestionAnswering[[transformers.AutoModelForQuestionAnswering]]

#### transformers.AutoModelForQuestionAnswering[[transformers.AutoModelForQuestionAnswering]]

```python
transformers.AutoModelForQuestionAnswering(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2355)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala penjawab pertanyaan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForQuestionAnswering.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [AlbertConfig](/docs/transformers/main/id/model_doc/albert#transformers.AlbertConfig): [AlbertForQuestionAnswering](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForQuestionAnswering) (model AlbertConfig) - Kelas konfigurasi [ArceeConfig](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeConfig): [ArceeForQuestionAnswering](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForQuestionAnswering) (model ArceeConfig) - Kelas konfigurasi [BartConfig](/docs/transformers/main/id/model_doc/bart#transformers.BartConfig): [BartForQuestionAnswering](/docs/transformers/main/id/model_doc/bart#transformers.BartForQuestionAnswering) (model BartConfig) - Kelas konfigurasi [BertConfig](/docs/transformers/main/id/model_doc/bert#transformers.BertConfig): [BertForQuestionAnswering](/docs/transformers/main/id/model_doc/bert#transformers.BertForQuestionAnswering) (model BertConfig) - Kelas konfigurasi [BigBirdConfig](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdConfig): [BigBirdForQuestionAnswering](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForQuestionAnswering) (model BigBirdConfig) - Kelas konfigurasi [BigBirdPegasusConfig](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusConfig): [BigBirdPegasusForQuestionAnswering](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForQuestionAnswering) (model BigBirdPegasusConfig) - Kelas konfigurasi [BloomConfig](/docs/transformers/main/id/model_doc/bloom#transformers.BloomConfig): [BloomForQuestionAnswering](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForQuestionAnswering) (model BloomConfig) - Kelas konfigurasi [CamembertConfig](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertConfig): [CamembertForQuestionAnswering](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForQuestionAnswering) (model CamembertConfig) - Kelas konfigurasi [CanineConfig](/docs/transformers/main/id/model_doc/canine#transformers.CanineConfig): [CanineForQuestionAnswering](/docs/transformers/main/id/model_doc/canine#transformers.CanineForQuestionAnswering) (model CanineConfig) - Kelas konfigurasi [ConvBertConfig](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertConfig): [ConvBertForQuestionAnswering](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForQuestionAnswering) (model ConvBertConfig) - Kelas konfigurasi [Data2VecTextConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextConfig): [Data2VecTextForQuestionAnswering](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForQuestionAnswering) (model Data2VecTextConfig) - Kelas konfigurasi [DebertaConfig](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaConfig): [DebertaForQuestionAnswering](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForQuestionAnswering) (model DebertaConfig) - Kelas konfigurasi [DebertaV2Config](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2Config): [DebertaV2ForQuestionAnswering](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForQuestionAnswering) (model DebertaV2Config) - Kelas konfigurasi [DiffLlamaConfig](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaConfig): [DiffLlamaForQuestionAnswering](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForQuestionAnswering) (model DiffLlamaConfig) - Kelas konfigurasi [DistilBertConfig](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertConfig): [DistilBertForQuestionAnswering](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForQuestionAnswering) (model DistilBertConfig) - Kelas konfigurasi [ElectraConfig](/docs/transformers/main/id/model_doc/electra#transformers.ElectraConfig): [ElectraForQuestionAnswering](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForQuestionAnswering) (model ElectraConfig) - Kelas konfigurasi [ErnieConfig](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieConfig): [ErnieForQuestionAnswering](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForQuestionAnswering) (model ErnieConfig) - Kelas konfigurasi [Exaone4Config](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4Config): [Exaone4ForQuestionAnswering](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForQuestionAnswering) (model Exaone4Config) - Kelas konfigurasi [FNetConfig](/docs/transformers/main/id/model_doc/fnet#transformers.FNetConfig): [FNetForQuestionAnswering](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForQuestionAnswering) (model FNetConfig) - Kelas konfigurasi [FalconConfig](/docs/transformers/main/id/model_doc/falcon#transformers.FalconConfig): [FalconForQuestionAnswering](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForQuestionAnswering) (model FalconConfig) - Kelas konfigurasi [FlaubertConfig](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertConfig): [FlaubertForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForQuestionAnsweringSimple) (model FlaubertConfig) - Kelas konfigurasi [FunnelConfig](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelConfig): [FunnelForQuestionAnswering](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForQuestionAnswering) (model FunnelConfig) - Kelas konfigurasi [GPT2Config](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2Config): [GPT2ForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForQuestionAnswering) (model GPT2Config) - Kelas konfigurasi [GPTJConfig](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJConfig): [GPTJForQuestionAnswering](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForQuestionAnswering) (model GPTJConfig) - Kelas konfigurasi [GPTNeoConfig](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoConfig): [GPTNeoForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForQuestionAnswering) (model GPTNeoConfig) - Kelas konfigurasi [GPTNeoXConfig](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXConfig): [GPTNeoXForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForQuestionAnswering) (model GPTNeoXConfig) - Kelas konfigurasi [IBertConfig](/docs/transformers/main/id/model_doc/ibert#transformers.IBertConfig): [IBertForQuestionAnswering](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForQuestionAnswering) (model IBertConfig) - Kelas konfigurasi [JinaEmbeddingsV3Config](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3Config): [JinaEmbeddingsV3ForQuestionAnswering](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForQuestionAnswering) (model JinaEmbeddingsV3Config) - Kelas konfigurasi [LEDConfig](/docs/transformers/main/id/model_doc/led#transformers.LEDConfig): [LEDForQuestionAnswering](/docs/transformers/main/id/model_doc/led#transformers.LEDForQuestionAnswering) (model LEDConfig) - Kelas konfigurasi [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config): [LayoutLMv2ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForQuestionAnswering) (model LayoutLMv2Config) - Kelas konfigurasi [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config): [LayoutLMv3ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForQuestionAnswering) (model LayoutLMv3Config) - Kelas konfigurasi [LiltConfig](/docs/transformers/main/id/model_doc/lilt#transformers.LiltConfig): [LiltForQuestionAnswering](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForQuestionAnswering) (model LiltConfig) - Kelas konfigurasi [LlamaConfig](/docs/transformers/main/id/model_doc/llama2#transformers.LlamaConfig): [LlamaForQuestionAnswering](/docs/transformers/main/id/model_doc/llama#transformers.LlamaForQuestionAnswering) (model LlamaConfig) - Kelas konfigurasi [LongformerConfig](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerConfig): [LongformerForQuestionAnswering](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForQuestionAnswering) (model LongformerConfig) - Kelas konfigurasi [LukeConfig](/docs/transformers/main/id/model_doc/luke#transformers.LukeConfig): [LukeForQuestionAnswering](/docs/transformers/main/id/model_doc/luke#transformers.LukeForQuestionAnswering) (model LukeConfig) - Kelas konfigurasi [LxmertConfig](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertConfig): [LxmertForQuestionAnswering](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertForQuestionAnswering) (model LxmertConfig) - Kelas konfigurasi [MBartConfig](/docs/transformers/main/id/model_doc/mbart#transformers.MBartConfig): [MBartForQuestionAnswering](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForQuestionAnswering) (model MBartConfig) - Kelas konfigurasi [MPNetConfig](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetConfig): [MPNetForQuestionAnswering](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForQuestionAnswering) (model MPNetConfig) - Kelas konfigurasi [MT5Config](/docs/transformers/main/id/model_doc/mt5#transformers.MT5Config): [MT5ForQuestionAnswering](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForQuestionAnswering) (model MT5Config) - Kelas konfigurasi [MarkupLMConfig](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMConfig): [MarkupLMForQuestionAnswering](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForQuestionAnswering) (model MarkupLMConfig) - Kelas konfigurasi [MegatronBertConfig](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertConfig): [MegatronBertForQuestionAnswering](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForQuestionAnswering) (model MegatronBertConfig) - Kelas konfigurasi [MiniMaxConfig](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxConfig): [MiniMaxForQuestionAnswering](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForQuestionAnswering) (model MiniMaxConfig) - Kelas konfigurasi [Ministral3Config](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3Config): [Ministral3ForQuestionAnswering](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForQuestionAnswering) (model Ministral3Config) - Kelas konfigurasi [MinistralConfig](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralConfig): [MinistralForQuestionAnswering](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForQuestionAnswering) (model MinistralConfig) - Kelas konfigurasi [MistralConfig](/docs/transformers/main/id/model_doc/mistral#transformers.MistralConfig): [MistralForQuestionAnswering](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForQuestionAnswering) (model MistralConfig) - Kelas konfigurasi [MixtralConfig](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralConfig): [MixtralForQuestionAnswering](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForQuestionAnswering) (model MixtralConfig) - Kelas konfigurasi [MobileBertConfig](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertConfig): [MobileBertForQuestionAnswering](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForQuestionAnswering) (model MobileBertConfig) - Kelas konfigurasi [ModernBertConfig](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertConfig): [ModernBertForQuestionAnswering](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForQuestionAnswering) (model ModernBertConfig) - Kelas konfigurasi [MptConfig](/docs/transformers/main/id/model_doc/mpt#transformers.MptConfig): [MptForQuestionAnswering](/docs/transformers/main/id/model_doc/mpt#transformers.MptForQuestionAnswering) (model MptConfig) - Kelas konfigurasi [MraConfig](/docs/transformers/main/id/model_doc/mra#transformers.MraConfig): [MraForQuestionAnswering](/docs/transformers/main/id/model_doc/mra#transformers.MraForQuestionAnswering) (model MraConfig) - Kelas konfigurasi [MvpConfig](/docs/transformers/main/id/model_doc/mvp#transformers.MvpConfig): [MvpForQuestionAnswering](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForQuestionAnswering) (model MvpConfig) - Kelas konfigurasi [NemotronConfig](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronConfig): [NemotronForQuestionAnswering](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForQuestionAnswering) (model NemotronConfig) - Kelas konfigurasi [NystromformerConfig](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerConfig): [NystromformerForQuestionAnswering](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForQuestionAnswering) (model NystromformerConfig) - Kelas konfigurasi [OPTConfig](/docs/transformers/main/id/model_doc/opt#transformers.OPTConfig): [OPTForQuestionAnswering](/docs/transformers/main/id/model_doc/opt#transformers.OPTForQuestionAnswering) (model OPTConfig) - Kelas konfigurasi [Qwen2Config](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2Config): [Qwen2ForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForQuestionAnswering) (model Qwen2Config) - Kelas konfigurasi [Qwen2MoeConfig](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeConfig): [Qwen2MoeForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForQuestionAnswering) (model Qwen2MoeConfig) - Kelas konfigurasi [Qwen3Config](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3Config): [Qwen3ForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForQuestionAnswering) (model Qwen3Config) - Kelas konfigurasi [Qwen3MoeConfig](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeConfig): [Qwen3MoeForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForQuestionAnswering) (model Qwen3MoeConfig) - Kelas konfigurasi [Qwen3NextConfig](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextConfig): [Qwen3NextForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForQuestionAnswering) (model Qwen3NextConfig) - Kelas konfigurasi [ReformerConfig](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerConfig): [ReformerForQuestionAnswering](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForQuestionAnswering) (model ReformerConfig) - Kelas konfigurasi [RemBertConfig](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertConfig): [RemBertForQuestionAnswering](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForQuestionAnswering) (model RemBertConfig) - Kelas konfigurasi [RoCBertConfig](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertConfig): [RoCBertForQuestionAnswering](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForQuestionAnswering) (model RoCBertConfig) - Kelas konfigurasi [RoFormerConfig](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerConfig): [RoFormerForQuestionAnswering](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForQuestionAnswering) (model RoFormerConfig) - Kelas konfigurasi [RobertaConfig](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaConfig): [RobertaForQuestionAnswering](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForQuestionAnswering) (model RobertaConfig) - Kelas konfigurasi [RobertaPreLayerNormConfig](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormConfig): [RobertaPreLayerNormForQuestionAnswering](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForQuestionAnswering) (model RobertaPreLayerNormConfig) - Kelas konfigurasi [SeedOssConfig](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssConfig): [SeedOssForQuestionAnswering](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForQuestionAnswering) (model SeedOssConfig) - Kelas konfigurasi [SmolLM3Config](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3Config): [SmolLM3ForQuestionAnswering](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForQuestionAnswering) (model SmolLM3Config) - Kelas konfigurasi [SplinterConfig](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterConfig): [SplinterForQuestionAnswering](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterForQuestionAnswering) (model SplinterConfig) - Kelas konfigurasi [SqueezeBertConfig](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertConfig): [SqueezeBertForQuestionAnswering](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForQuestionAnswering) (model SqueezeBertConfig) - Kelas konfigurasi [T5Config](/docs/transformers/main/id/model_doc/t5#transformers.T5Config): [T5ForQuestionAnswering](/docs/transformers/main/id/model_doc/t5#transformers.T5ForQuestionAnswering) (model T5Config) - Kelas konfigurasi [UMT5Config](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5Config): [UMT5ForQuestionAnswering](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForQuestionAnswering) (model UMT5Config) - Kelas konfigurasi [UltrabertConfig](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertConfig): [UltrabertForQuestionAnswering](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForQuestionAnswering) (model UltrabertConfig) - Kelas konfigurasi [XLMConfig](/docs/transformers/main/id/model_doc/xlm#transformers.XLMConfig): [XLMForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForQuestionAnsweringSimple) (model XLMConfig) - Kelas konfigurasi [XLMRobertaConfig](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaConfig): [XLMRobertaForQuestionAnswering](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForQuestionAnswering) (model XLMRobertaConfig) - Kelas konfigurasi [XLMRobertaXLConfig](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLConfig): [XLMRobertaXLForQuestionAnswering](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForQuestionAnswering) (model XLMRobertaXLConfig) - Kelas konfigurasi [XLNetConfig](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetConfig): [XLNetForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForQuestionAnsweringSimple) (model XLNetConfig) - Kelas konfigurasi [XmodConfig](/docs/transformers/main/id/model_doc/xmod#transformers.XmodConfig): [XmodForQuestionAnswering](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForQuestionAnswering) (model XmodConfig) - Kelas konfigurasi [YosoConfig](/docs/transformers/main/id/model_doc/yoso#transformers.YosoConfig): [YosoForQuestionAnswering](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForQuestionAnswering) (model YosoConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala untuk menjawab pertanyaan) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForQuestionAnswering

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForQuestionAnswering.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForQuestionAnswering.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala penjawab pertanyaan) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **albert** -- [AlbertForQuestionAnswering](/docs/transformers/main/id/model_doc/albert#transformers.AlbertForQuestionAnswering) (model AlbertConfig)
- **arcee** -- [ArceeForQuestionAnswering](/docs/transformers/main/id/model_doc/arcee#transformers.ArceeForQuestionAnswering) (model ArceeConfig)
- **bart** -- [BartForQuestionAnswering](/docs/transformers/main/id/model_doc/bart#transformers.BartForQuestionAnswering) (model BartConfig)
- **bert** -- [BertForQuestionAnswering](/docs/transformers/main/id/model_doc/bert#transformers.BertForQuestionAnswering) (model BertConfig)
- **big_bird** -- [BigBirdForQuestionAnswering](/docs/transformers/main/id/model_doc/big_bird#transformers.BigBirdForQuestionAnswering) (model BigBirdConfig)
- **bigbird_pegasus** -- [BigBirdPegasusForQuestionAnswering](/docs/transformers/main/id/model_doc/bigbird_pegasus#transformers.BigBirdPegasusForQuestionAnswering) (model BigBirdPegasusConfig)
- **bloom** -- [BloomForQuestionAnswering](/docs/transformers/main/id/model_doc/bloom#transformers.BloomForQuestionAnswering) (model BloomConfig)
- **camembert** -- [CamembertForQuestionAnswering](/docs/transformers/main/id/model_doc/camembert#transformers.CamembertForQuestionAnswering) (model CamembertConfig)
- **canine** -- [CanineForQuestionAnswering](/docs/transformers/main/id/model_doc/canine#transformers.CanineForQuestionAnswering) (model CanineConfig)
- **convbert** -- [ConvBertForQuestionAnswering](/docs/transformers/main/id/model_doc/convbert#transformers.ConvBertForQuestionAnswering) (model ConvBertConfig)
- **data2vec-text** -- [Data2VecTextForQuestionAnswering](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecTextForQuestionAnswering) (model Data2VecTextConfig)
- **deberta** -- [DebertaForQuestionAnswering](/docs/transformers/main/id/model_doc/deberta#transformers.DebertaForQuestionAnswering) (model DebertaConfig)
- **deberta-v2** -- [DebertaV2ForQuestionAnswering](/docs/transformers/main/id/model_doc/deberta-v2#transformers.DebertaV2ForQuestionAnswering) (model DebertaV2Config)
- **diffllama** -- [DiffLlamaForQuestionAnswering](/docs/transformers/main/id/model_doc/diffllama#transformers.DiffLlamaForQuestionAnswering) (model DiffLlamaConfig)
- **distilbert** -- [DistilBertForQuestionAnswering](/docs/transformers/main/id/model_doc/distilbert#transformers.DistilBertForQuestionAnswering) (model DistilBertConfig)
- **electra** -- [ElectraForQuestionAnswering](/docs/transformers/main/id/model_doc/electra#transformers.ElectraForQuestionAnswering) (model ElectraConfig)
- **ernie** -- [ErnieForQuestionAnswering](/docs/transformers/main/id/model_doc/ernie#transformers.ErnieForQuestionAnswering) (model ErnieConfig)
- **exaone4** -- [Exaone4ForQuestionAnswering](/docs/transformers/main/id/model_doc/exaone4#transformers.Exaone4ForQuestionAnswering) (model Exaone4Config)
- **falcon** -- [FalconForQuestionAnswering](/docs/transformers/main/id/model_doc/falcon#transformers.FalconForQuestionAnswering) (model FalconConfig)
- **flaubert** -- [FlaubertForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/flaubert#transformers.FlaubertForQuestionAnsweringSimple) (model FlaubertConfig)
- **fnet** -- [FNetForQuestionAnswering](/docs/transformers/main/id/model_doc/fnet#transformers.FNetForQuestionAnswering) (model FNetConfig)
- **funnel** -- [FunnelForQuestionAnswering](/docs/transformers/main/id/model_doc/funnel#transformers.FunnelForQuestionAnswering) (model FunnelConfig)
- **gpt2** -- [GPT2ForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt2#transformers.GPT2ForQuestionAnswering) (model GPT2Config)
- **gpt_neo** -- [GPTNeoForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt_neo#transformers.GPTNeoForQuestionAnswering) (model GPTNeoConfig)
- **gpt_neox** -- [GPTNeoXForQuestionAnswering](/docs/transformers/main/id/model_doc/gpt_neox#transformers.GPTNeoXForQuestionAnswering) (model GPTNeoXConfig)
- **gptj** -- [GPTJForQuestionAnswering](/docs/transformers/main/id/model_doc/gptj#transformers.GPTJForQuestionAnswering) (model GPTJConfig)
- **ibert** -- [IBertForQuestionAnswering](/docs/transformers/main/id/model_doc/ibert#transformers.IBertForQuestionAnswering) (model IBertConfig)
- **jina_embeddings_v3** -- [JinaEmbeddingsV3ForQuestionAnswering](/docs/transformers/main/id/model_doc/jina_embeddings_v3#transformers.JinaEmbeddingsV3ForQuestionAnswering) (model JinaEmbeddingsV3Config)
- **layoutlmv2** -- [LayoutLMv2ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForQuestionAnswering) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForQuestionAnswering) (model LayoutLMv3Config)
- **led** -- [LEDForQuestionAnswering](/docs/transformers/main/id/model_doc/led#transformers.LEDForQuestionAnswering) (model LEDConfig)
- **lilt** -- [LiltForQuestionAnswering](/docs/transformers/main/id/model_doc/lilt#transformers.LiltForQuestionAnswering) (model LiltConfig)
- **llama** -- [LlamaForQuestionAnswering](/docs/transformers/main/id/model_doc/llama#transformers.LlamaForQuestionAnswering) (model LlamaConfig)
- **longformer** -- [LongformerForQuestionAnswering](/docs/transformers/main/id/model_doc/longformer#transformers.LongformerForQuestionAnswering) (model LongformerConfig)
- **luke** -- [LukeForQuestionAnswering](/docs/transformers/main/id/model_doc/luke#transformers.LukeForQuestionAnswering) (model LukeConfig)
- **lxmert** -- [LxmertForQuestionAnswering](/docs/transformers/main/id/model_doc/lxmert#transformers.LxmertForQuestionAnswering) (model LxmertConfig)
- **markuplm** -- [MarkupLMForQuestionAnswering](/docs/transformers/main/id/model_doc/markuplm#transformers.MarkupLMForQuestionAnswering) (model MarkupLMConfig)
- **mbart** -- [MBartForQuestionAnswering](/docs/transformers/main/id/model_doc/mbart#transformers.MBartForQuestionAnswering) (model MBartConfig)
- **megatron-bert** -- [MegatronBertForQuestionAnswering](/docs/transformers/main/id/model_doc/megatron-bert#transformers.MegatronBertForQuestionAnswering) (model MegatronBertConfig)
- **minimax** -- [MiniMaxForQuestionAnswering](/docs/transformers/main/id/model_doc/minimax#transformers.MiniMaxForQuestionAnswering) (model MiniMaxConfig)
- **ministral** -- [MinistralForQuestionAnswering](/docs/transformers/main/id/model_doc/ministral#transformers.MinistralForQuestionAnswering) (model MinistralConfig)
- **ministral3** -- [Ministral3ForQuestionAnswering](/docs/transformers/main/id/model_doc/ministral3#transformers.Ministral3ForQuestionAnswering) (model Ministral3Config)
- **mistral** -- [MistralForQuestionAnswering](/docs/transformers/main/id/model_doc/mistral#transformers.MistralForQuestionAnswering) (model MistralConfig)
- **mixtral** -- [MixtralForQuestionAnswering](/docs/transformers/main/id/model_doc/mixtral#transformers.MixtralForQuestionAnswering) (model MixtralConfig)
- **mobilebert** -- [MobileBertForQuestionAnswering](/docs/transformers/main/id/model_doc/mobilebert#transformers.MobileBertForQuestionAnswering) (model MobileBertConfig)
- **modernbert** -- [ModernBertForQuestionAnswering](/docs/transformers/main/id/model_doc/modernbert#transformers.ModernBertForQuestionAnswering) (model ModernBertConfig)
- **mpnet** -- [MPNetForQuestionAnswering](/docs/transformers/main/id/model_doc/mpnet#transformers.MPNetForQuestionAnswering) (model MPNetConfig)
- **mpt** -- [MptForQuestionAnswering](/docs/transformers/main/id/model_doc/mpt#transformers.MptForQuestionAnswering) (model MptConfig)
- **mra** -- [MraForQuestionAnswering](/docs/transformers/main/id/model_doc/mra#transformers.MraForQuestionAnswering) (model MraConfig)
- **mt5** -- [MT5ForQuestionAnswering](/docs/transformers/main/id/model_doc/mt5#transformers.MT5ForQuestionAnswering) (model MT5Config)
- **mvp** -- [MvpForQuestionAnswering](/docs/transformers/main/id/model_doc/mvp#transformers.MvpForQuestionAnswering) (model MvpConfig)
- **nemotron** -- [NemotronForQuestionAnswering](/docs/transformers/main/id/model_doc/nemotron#transformers.NemotronForQuestionAnswering) (model NemotronConfig)
- **nystromformer** -- [NystromformerForQuestionAnswering](/docs/transformers/main/id/model_doc/nystromformer#transformers.NystromformerForQuestionAnswering) (model NystromformerConfig)
- **opt** -- [OPTForQuestionAnswering](/docs/transformers/main/id/model_doc/opt#transformers.OPTForQuestionAnswering) (model OPTConfig)
- **qwen2** -- [Qwen2ForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen2#transformers.Qwen2ForQuestionAnswering) (model Qwen2Config)
- **qwen2_moe** -- [Qwen2MoeForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen2_moe#transformers.Qwen2MoeForQuestionAnswering) (model Qwen2MoeConfig)
- **qwen3** -- [Qwen3ForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3#transformers.Qwen3ForQuestionAnswering) (model Qwen3Config)
- **qwen3_moe** -- [Qwen3MoeForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3_moe#transformers.Qwen3MoeForQuestionAnswering) (model Qwen3MoeConfig)
- **qwen3_next** -- [Qwen3NextForQuestionAnswering](/docs/transformers/main/id/model_doc/qwen3_next#transformers.Qwen3NextForQuestionAnswering) (model Qwen3NextConfig)
- **reformer** -- [ReformerForQuestionAnswering](/docs/transformers/main/id/model_doc/reformer#transformers.ReformerForQuestionAnswering) (model ReformerConfig)
- **rembert** -- [RemBertForQuestionAnswering](/docs/transformers/main/id/model_doc/rembert#transformers.RemBertForQuestionAnswering) (model RemBertConfig)
- **roberta** -- [RobertaForQuestionAnswering](/docs/transformers/main/id/model_doc/roberta#transformers.RobertaForQuestionAnswering) (model RobertaConfig)
- **roberta-prelayernorm** -- [RobertaPreLayerNormForQuestionAnswering](/docs/transformers/main/id/model_doc/roberta-prelayernorm#transformers.RobertaPreLayerNormForQuestionAnswering) (model RobertaPreLayerNormConfig)
- **roc_bert** -- [RoCBertForQuestionAnswering](/docs/transformers/main/id/model_doc/roc_bert#transformers.RoCBertForQuestionAnswering) (model RoCBertConfig)
- **roformer** -- [RoFormerForQuestionAnswering](/docs/transformers/main/id/model_doc/roformer#transformers.RoFormerForQuestionAnswering) (model RoFormerConfig)
- **seed_oss** -- [SeedOssForQuestionAnswering](/docs/transformers/main/id/model_doc/seed_oss#transformers.SeedOssForQuestionAnswering) (model SeedOssConfig)
- **smollm3** -- [SmolLM3ForQuestionAnswering](/docs/transformers/main/id/model_doc/smollm3#transformers.SmolLM3ForQuestionAnswering) (model SmolLM3Config)
- **splinter** -- [SplinterForQuestionAnswering](/docs/transformers/main/id/model_doc/splinter#transformers.SplinterForQuestionAnswering) (model SplinterConfig)
- **squeezebert** -- [SqueezeBertForQuestionAnswering](/docs/transformers/main/id/model_doc/squeezebert#transformers.SqueezeBertForQuestionAnswering) (model SqueezeBertConfig)
- **t5** -- [T5ForQuestionAnswering](/docs/transformers/main/id/model_doc/t5#transformers.T5ForQuestionAnswering) (model T5Config)
- **ultrabert** -- [UltrabertForQuestionAnswering](/docs/transformers/main/id/model_doc/ultrabert#transformers.UltrabertForQuestionAnswering) (model UltrabertConfig)
- **umt5** -- [UMT5ForQuestionAnswering](/docs/transformers/main/id/model_doc/umt5#transformers.UMT5ForQuestionAnswering) (model UMT5Config)
- **xlm** -- [XLMForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/xlm#transformers.XLMForQuestionAnsweringSimple) (model XLMConfig)
- **xlm-roberta** -- [XLMRobertaForQuestionAnswering](/docs/transformers/main/id/model_doc/xlm-roberta#transformers.XLMRobertaForQuestionAnswering) (model XLMRobertaConfig)
- **xlm-roberta-xl** -- [XLMRobertaXLForQuestionAnswering](/docs/transformers/main/id/model_doc/xlm-roberta-xl#transformers.XLMRobertaXLForQuestionAnswering) (model XLMRobertaXLConfig)
- **xlnet** -- [XLNetForQuestionAnsweringSimple](/docs/transformers/main/id/model_doc/xlnet#transformers.XLNetForQuestionAnsweringSimple) (model XLNetConfig)
- **xmod** -- [XmodForQuestionAnswering](/docs/transformers/main/id/model_doc/xmod#transformers.XmodForQuestionAnswering) (model XmodConfig)
- **yoso** -- [YosoForQuestionAnswering](/docs/transformers/main/id/model_doc/yoso#transformers.YosoForQuestionAnswering) (model YosoConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForQuestionAnswering

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForQuestionAnswering.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForQuestionAnswering.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTextEncoding[[transformers.AutoModelForTextEncoding]]

#### transformers.AutoModelForTextEncoding[[transformers.AutoModelForTextEncoding]]

```python
transformers.AutoModelForTextEncoding(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2289)

## Visi komputer

Kelas auto berikut tersedia untuk tugas-tugas computer vision berikut.

### AutoModelForDepthEstimation[[transformers.AutoModelForDepthEstimation]]

#### transformers.AutoModelForDepthEstimation[[transformers.AutoModelForDepthEstimation]]

```python
transformers.AutoModelForDepthEstimation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2493)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala estimasi kedalaman) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForDepthEstimation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [CHMv2Config](/docs/transformers/main/id/model_doc/chmv2#transformers.CHMv2Config): [CHMv2ForDepthEstimation](/docs/transformers/main/id/model_doc/chmv2#transformers.CHMv2ForDepthEstimation) (model CHMv2Config) - Kelas konfigurasi [DPTConfig](/docs/transformers/main/id/model_doc/dpt#transformers.DPTConfig): [DPTForDepthEstimation](/docs/transformers/main/id/model_doc/dpt#transformers.DPTForDepthEstimation) (model DPTConfig) - Kelas konfigurasi [DepthAnythingConfig](/docs/transformers/main/id/model_doc/depth_anything#transformers.DepthAnythingConfig): [DepthAnythingForDepthEstimation](/docs/transformers/main/id/model_doc/depth_anything#transformers.DepthAnythingForDepthEstimation) (model DepthAnythingConfig) - Kelas konfigurasi [DepthProConfig](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProConfig): [DepthProForDepthEstimation](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProForDepthEstimation) (model DepthProConfig) - Kelas konfigurasi [GLPNConfig](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNConfig): [GLPNForDepthEstimation](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNForDepthEstimation) (model GLPNConfig) - Kelas konfigurasi [PromptDepthAnythingConfig](/docs/transformers/main/id/model_doc/prompt_depth_anything#transformers.PromptDepthAnythingConfig): [PromptDepthAnythingForDepthEstimation](/docs/transformers/main/id/model_doc/prompt_depth_anything#transformers.PromptDepthAnythingForDepthEstimation) (model PromptDepthAnythingConfig) - Kelas konfigurasi [Tipsv2DptConfig](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptConfig): [Tipsv2DptForDepthEstimation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForDepthEstimation) (model Tipsv2DptConfig) - Kelas konfigurasi [ZoeDepthConfig](/docs/transformers/main/id/model_doc/zoedepth#transformers.ZoeDepthConfig): [ZoeDepthForDepthEstimation](/docs/transformers/main/id/model_doc/zoedepth#transformers.ZoeDepthForDepthEstimation) (model ZoeDepthConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala estimasi kedalaman) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForDepthEstimation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForDepthEstimation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForDepthEstimation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala estimasi kedalaman) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **chmv2** -- [CHMv2ForDepthEstimation](/docs/transformers/main/id/model_doc/chmv2#transformers.CHMv2ForDepthEstimation) (model CHMv2Config)
- **depth_anything** -- [DepthAnythingForDepthEstimation](/docs/transformers/main/id/model_doc/depth_anything#transformers.DepthAnythingForDepthEstimation) (model DepthAnythingConfig)
- **depth_pro** -- [DepthProForDepthEstimation](/docs/transformers/main/id/model_doc/depth_pro#transformers.DepthProForDepthEstimation) (model DepthProConfig)
- **dpt** -- [DPTForDepthEstimation](/docs/transformers/main/id/model_doc/dpt#transformers.DPTForDepthEstimation) (model DPTConfig)
- **glpn** -- [GLPNForDepthEstimation](/docs/transformers/main/id/model_doc/glpn#transformers.GLPNForDepthEstimation) (model GLPNConfig)
- **prompt_depth_anything** -- [PromptDepthAnythingForDepthEstimation](/docs/transformers/main/id/model_doc/prompt_depth_anything#transformers.PromptDepthAnythingForDepthEstimation) (model PromptDepthAnythingConfig)
- **tipsv2_dpt** -- [Tipsv2DptForDepthEstimation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForDepthEstimation) (model Tipsv2DptConfig)
- **zoedepth** -- [ZoeDepthForDepthEstimation](/docs/transformers/main/id/model_doc/zoedepth#transformers.ZoeDepthForDepthEstimation) (model ZoeDepthConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForDepthEstimation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForDepthEstimation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForDepthEstimation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForNormalEstimation[[transformers.AutoModelForNormalEstimation]]

#### transformers.AutoModelForNormalEstimation[[transformers.AutoModelForNormalEstimation]]

```python
transformers.AutoModelForNormalEstimation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2633)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala estimasi normal) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForNormalEstimation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config): [Sapiens2ForNormalEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForNormalEstimation) (model Sapiens2Config) - Kelas konfigurasi [Tipsv2DptConfig](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptConfig): [Tipsv2DptForNormalEstimation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForNormalEstimation) (model Tipsv2DptConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala estimasi normal) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForNormalEstimation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForNormalEstimation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForNormalEstimation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala estimasi normal) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **sapiens2** -- [Sapiens2ForNormalEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForNormalEstimation) (model Sapiens2Config)
- **tipsv2_dpt** -- [Tipsv2DptForNormalEstimation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForNormalEstimation) (model Tipsv2DptConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForNormalEstimation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForNormalEstimation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForNormalEstimation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForPointmapEstimation[[transformers.AutoModelForPointmapEstimation]]

#### transformers.AutoModelForPointmapEstimation[[transformers.AutoModelForPointmapEstimation]]

```python
transformers.AutoModelForPointmapEstimation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2640)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala estimasi pointmap) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForPointmapEstimation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config): [Sapiens2ForPointmapEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForPointmapEstimation) (model Sapiens2Config)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala estimasi pointmap) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPointmapEstimation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForPointmapEstimation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForPointmapEstimation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala estimasi pointmap) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **sapiens2** -- [Sapiens2ForPointmapEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForPointmapEstimation) (model Sapiens2Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPointmapEstimation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForPointmapEstimation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForPointmapEstimation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForImageMatting[[transformers.AutoModelForImageMatting]]

#### transformers.AutoModelForImageMatting[[transformers.AutoModelForImageMatting]]

```python
transformers.AutoModelForImageMatting(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2626)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala matting gambar) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForImageMatting.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config): [Sapiens2ForImageMatting](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForImageMatting) (model Sapiens2Config) - Kelas konfigurasi [VitMatteConfig](/docs/transformers/main/id/model_doc/vitmatte#transformers.VitMatteConfig): [VitMatteForImageMatting](/docs/transformers/main/id/model_doc/vitmatte#transformers.VitMatteForImageMatting) (model VitMatteConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala matting gambar) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageMatting

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForImageMatting.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForImageMatting.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala matting gambar) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **sapiens2** -- [Sapiens2ForImageMatting](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForImageMatting) (model Sapiens2Config)
- **vitmatte** -- [VitMatteForImageMatting](/docs/transformers/main/id/model_doc/vitmatte#transformers.VitMatteForImageMatting) (model VitMatteConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageMatting

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForImageMatting.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForImageMatting.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTextRecognition[[transformers.AutoModelForTextRecognition]]

#### transformers.AutoModelForTextRecognition[[transformers.AutoModelForTextRecognition]]

```python
transformers.AutoModelForTextRecognition(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2500)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pengenalan teks) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTextRecognition.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [PPOCRV5MobileRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecConfig): [PPOCRV5MobileRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecForTextRecognition) (model PPOCRV5MobileRecConfig) - Kelas konfigurasi [PPOCRV5ServerRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecConfig): [PPOCRV5ServerRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecForTextRecognition) (model PPOCRV5ServerRecConfig) - Kelas konfigurasi [PPOCRV6SmallRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecConfig): [PPOCRV6SmallRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecForTextRecognition) (model PPOCRV6SmallRecConfig) - Kelas konfigurasi [PPOCRV6TinyRecConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecConfig): [PPOCRV6TinyRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecForTextRecognition) (model PPOCRV6TinyRecConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pengenalan teks) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTextRecognition

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForTextRecognition.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTextRecognition.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pengenalan teks) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **pp_ocrv5_mobile_rec** -- [PPOCRV5MobileRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_rec#transformers.PPOCRV5MobileRecForTextRecognition) (model PPOCRV5MobileRecConfig)
- **pp_ocrv5_server_rec** -- [PPOCRV5ServerRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv5_server_rec#transformers.PPOCRV5ServerRecForTextRecognition) (model PPOCRV5ServerRecConfig)
- **pp_ocrv6_small_rec** -- [PPOCRV6SmallRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv6_small_rec#transformers.PPOCRV6SmallRecForTextRecognition) (model PPOCRV6SmallRecConfig)
- **pp_ocrv6_tiny_rec** -- [PPOCRV6TinyRecForTextRecognition](/docs/transformers/main/id/model_doc/pp_ocrv6_tiny_rec#transformers.PPOCRV6TinyRecForTextRecognition) (model PPOCRV6TinyRecConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTextRecognition

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTextRecognition.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForTextRecognition.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTableRecognition[[transformers.AutoModelForTableRecognition]]

#### transformers.AutoModelForTableRecognition[[transformers.AutoModelForTableRecognition]]

```python
transformers.AutoModelForTableRecognition(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2507)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pengenalan tabel) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTableRecognition.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [SLANeXtConfig](/docs/transformers/main/id/model_doc/slanext#transformers.SLANeXtConfig): [SLANeXtForTableRecognition](/docs/transformers/main/id/model_doc/slanext#transformers.SLANeXtForTableRecognition) (model SLANeXtConfig) - Kelas konfigurasi [SLANetConfig](/docs/transformers/main/id/model_doc/slanet#transformers.SLANetConfig): [SLANetForTableRecognition](/docs/transformers/main/id/model_doc/slanet#transformers.SLANetForTableRecognition) (model SLANetConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pengenalan tabel) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTableRecognition

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForTableRecognition.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTableRecognition.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pengenalan tabel) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **slanet** -- [SLANetForTableRecognition](/docs/transformers/main/id/model_doc/slanet#transformers.SLANetForTableRecognition) (model SLANetConfig)
- **slanext** -- [SLANeXtForTableRecognition](/docs/transformers/main/id/model_doc/slanext#transformers.SLANeXtForTableRecognition) (model SLANeXtConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTableRecognition

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTableRecognition.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForTableRecognition.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForImageClassification[[transformers.AutoModelForImageClassification]]

#### transformers.AutoModelForImageClassification[[transformers.AutoModelForImageClassification]]

```python
transformers.AutoModelForImageClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2418)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForImageClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [BeitConfig](/docs/transformers/main/id/model_doc/beit#transformers.BeitConfig): [BeitForImageClassification](/docs/transformers/main/id/model_doc/beit#transformers.BeitForImageClassification) (model BeitConfig) - Kelas konfigurasi [BitConfig](/docs/transformers/main/id/model_doc/bit#transformers.BitConfig): [BitForImageClassification](/docs/transformers/main/id/model_doc/bit#transformers.BitForImageClassification) (model BitConfig) - Kelas konfigurasi [CLIPConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPConfig): [CLIPForImageClassification](/docs/transformers/main/id/model_doc/clip#transformers.CLIPForImageClassification) (model CLIPConfig) - Kelas konfigurasi [ConvNextConfig](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextConfig): [ConvNextForImageClassification](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextForImageClassification) (model ConvNextConfig) - Kelas konfigurasi [ConvNextV2Config](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2Config): [ConvNextV2ForImageClassification](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2ForImageClassification) (model ConvNextV2Config) - Kelas konfigurasi [CvtConfig](/docs/transformers/main/id/model_doc/cvt#transformers.CvtConfig): [CvtForImageClassification](/docs/transformers/main/id/model_doc/cvt#transformers.CvtForImageClassification) (model CvtConfig) - Kelas konfigurasi [Data2VecVisionConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionConfig): [Data2VecVisionForImageClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionForImageClassification) (model Data2VecVisionConfig) - Kelas konfigurasi [DeiTConfig](/docs/transformers/main/id/model_doc/deit#transformers.DeiTConfig): [DeiTForImageClassification](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForImageClassification) atau [DeiTForImageClassificationWithTeacher](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForImageClassificationWithTeacher) (model DeiTConfig) - Kelas konfigurasi [DinatConfig](/docs/transformers/main/id/model_doc/dinat#transformers.DinatConfig): [DinatForImageClassification](/docs/transformers/main/id/model_doc/dinat#transformers.DinatForImageClassification) (model DinatConfig) - Kelas konfigurasi [Dinov2Config](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2Config): [Dinov2ForImageClassification](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2ForImageClassification) (model Dinov2Config) - Kelas konfigurasi [Dinov2WithRegistersConfig](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersConfig): [Dinov2WithRegistersForImageClassification](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersForImageClassification) (model Dinov2WithRegistersConfig) - Kelas konfigurasi [DonutSwinConfig](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinConfig): [DonutSwinForImageClassification](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinForImageClassification) (model DonutSwinConfig) - Kelas konfigurasi [EfficientNetConfig](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetConfig): [EfficientNetForImageClassification](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetForImageClassification) (model EfficientNetConfig) - Kelas konfigurasi [FocalNetConfig](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetConfig): [FocalNetForImageClassification](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetForImageClassification) (model FocalNetConfig) - Kelas konfigurasi [HGNetV2Config](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2Config): [HGNetV2ForImageClassification](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2ForImageClassification) (model HGNetV2Config) - Kelas konfigurasi [HieraConfig](/docs/transformers/main/id/model_doc/hiera#transformers.HieraConfig): [HieraForImageClassification](/docs/transformers/main/id/model_doc/hiera#transformers.HieraForImageClassification) (model HieraConfig) - Kelas konfigurasi [IJepaConfig](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaConfig): [IJepaForImageClassification](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaForImageClassification) (model IJepaConfig) - Kelas konfigurasi [ImageGPTConfig](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTConfig): [ImageGPTForImageClassification](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTForImageClassification) (model ImageGPTConfig) - Kelas konfigurasi [LevitConfig](/docs/transformers/main/id/model_doc/levit#transformers.LevitConfig): [LevitForImageClassification](/docs/transformers/main/id/model_doc/levit#transformers.LevitForImageClassification) atau [LevitForImageClassificationWithTeacher](/docs/transformers/main/id/model_doc/levit#transformers.LevitForImageClassificationWithTeacher) (model LevitConfig) - Kelas konfigurasi [MetaClip2Config](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Config): [MetaClip2ForImageClassification](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2ForImageClassification) (model MetaClip2Config) - Kelas konfigurasi [MobileNetV1Config](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1Config): [MobileNetV1ForImageClassification](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1ForImageClassification) (model MobileNetV1Config) - Kelas konfigurasi [MobileNetV2Config](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Config): [MobileNetV2ForImageClassification](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2ForImageClassification) (model MobileNetV2Config) - Kelas konfigurasi [MobileViTConfig](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTConfig): [MobileViTForImageClassification](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTForImageClassification) (model MobileViTConfig) - Kelas konfigurasi [MobileViTV2Config](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Config): [MobileViTV2ForImageClassification](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2ForImageClassification) (model MobileViTV2Config) - Kelas konfigurasi [PPLCNetConfig](/docs/transformers/main/id/model_doc/pp_lcnet#transformers.PPLCNetConfig): [PPLCNetForImageClassification](/docs/transformers/main/id/model_doc/pp_lcnet#transformers.PPLCNetForImageClassification) (model PPLCNetConfig) - Kelas konfigurasi [PerceiverConfig](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverConfig): [PerceiverForImageClassificationLearned](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationLearned) atau [PerceiverForImageClassificationFourier](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationFourier) atau [PerceiverForImageClassificationConvProcessing](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationConvProcessing) (model PerceiverConfig) - Kelas konfigurasi [PoolFormerConfig](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerConfig): [PoolFormerForImageClassification](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerForImageClassification) (model PoolFormerConfig) - Kelas konfigurasi [PvtConfig](/docs/transformers/main/id/model_doc/pvt#transformers.PvtConfig): [PvtForImageClassification](/docs/transformers/main/id/model_doc/pvt#transformers.PvtForImageClassification) (model PvtConfig) - Kelas konfigurasi [PvtV2Config](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2Config): [PvtV2ForImageClassification](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2ForImageClassification) (model PvtV2Config) - Kelas konfigurasi [RegNetConfig](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetConfig): [RegNetForImageClassification](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetForImageClassification) (model RegNetConfig) - Kelas konfigurasi [ResNetConfig](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetConfig): [ResNetForImageClassification](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetForImageClassification) (model ResNetConfig) - Kelas konfigurasi [SegformerConfig](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerConfig): [SegformerForImageClassification](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerForImageClassification) (model SegformerConfig) - Kelas konfigurasi [ShieldGemma2Config](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2Config): [ShieldGemma2ForImageClassification](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2ForImageClassification) (model ShieldGemma2Config) - Kelas konfigurasi [Siglip2Config](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Config): [Siglip2ForImageClassification](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2ForImageClassification) (model Siglip2Config) - Kelas konfigurasi [SiglipConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipConfig): [SiglipForImageClassification](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipForImageClassification) (model SiglipConfig) - Kelas konfigurasi [SwiftFormerConfig](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerConfig): [SwiftFormerForImageClassification](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerForImageClassification) (model SwiftFormerConfig) - Kelas konfigurasi [SwinConfig](/docs/transformers/main/id/model_doc/swin#transformers.SwinConfig): [SwinForImageClassification](/docs/transformers/main/id/model_doc/swin#transformers.SwinForImageClassification) (model SwinConfig) - Kelas konfigurasi [Swinv2Config](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Config): [Swinv2ForImageClassification](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2ForImageClassification) (model Swinv2Config) - Kelas konfigurasi [TextNetConfig](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetConfig): [TextNetForImageClassification](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetForImageClassification) (model TextNetConfig) - Kelas konfigurasi [TimmWrapperConfig](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperConfig): [TimmWrapperForImageClassification](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperForImageClassification) (model TimmWrapperConfig) - Kelas konfigurasi [ViTConfig](/docs/transformers/main/id/model_doc/vit#transformers.ViTConfig): [ViTForImageClassification](/docs/transformers/main/id/model_doc/vit#transformers.ViTForImageClassification) (model ViTConfig) - Kelas konfigurasi [ViTMSNConfig](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNConfig): [ViTMSNForImageClassification](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNForImageClassification) (model ViTMSNConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForImageClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForImageClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **beit** -- [BeitForImageClassification](/docs/transformers/main/id/model_doc/beit#transformers.BeitForImageClassification) (model BeitConfig)
- **bit** -- [BitForImageClassification](/docs/transformers/main/id/model_doc/bit#transformers.BitForImageClassification) (model BitConfig)
- **clip** -- [CLIPForImageClassification](/docs/transformers/main/id/model_doc/clip#transformers.CLIPForImageClassification) (model CLIPConfig)
- **convnext** -- [ConvNextForImageClassification](/docs/transformers/main/id/model_doc/convnext#transformers.ConvNextForImageClassification) (model ConvNextConfig)
- **convnextv2** -- [ConvNextV2ForImageClassification](/docs/transformers/main/id/model_doc/convnextv2#transformers.ConvNextV2ForImageClassification) (model ConvNextV2Config)
- **cvt** -- [CvtForImageClassification](/docs/transformers/main/id/model_doc/cvt#transformers.CvtForImageClassification) (model CvtConfig)
- **data2vec-vision** -- [Data2VecVisionForImageClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionForImageClassification) (model Data2VecVisionConfig)
- **deit** -- [DeiTForImageClassification](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForImageClassification) atau [DeiTForImageClassificationWithTeacher](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForImageClassificationWithTeacher) (model DeiTConfig)
- **dinat** -- [DinatForImageClassification](/docs/transformers/main/id/model_doc/dinat#transformers.DinatForImageClassification) (model DinatConfig)
- **dinov2** -- [Dinov2ForImageClassification](/docs/transformers/main/id/model_doc/dinov2#transformers.Dinov2ForImageClassification) (model Dinov2Config)
- **dinov2_with_registers** -- [Dinov2WithRegistersForImageClassification](/docs/transformers/main/id/model_doc/dinov2_with_registers#transformers.Dinov2WithRegistersForImageClassification) (model Dinov2WithRegistersConfig)
- **donut-swin** -- [DonutSwinForImageClassification](/docs/transformers/main/id/model_doc/donut#transformers.DonutSwinForImageClassification) (model DonutSwinConfig)
- **efficientnet** -- [EfficientNetForImageClassification](/docs/transformers/main/id/model_doc/efficientnet#transformers.EfficientNetForImageClassification) (model EfficientNetConfig)
- **focalnet** -- [FocalNetForImageClassification](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetForImageClassification) (model FocalNetConfig)
- **hgnet_v2** -- [HGNetV2ForImageClassification](/docs/transformers/main/id/model_doc/hgnet_v2#transformers.HGNetV2ForImageClassification) (model HGNetV2Config)
- **hiera** -- [HieraForImageClassification](/docs/transformers/main/id/model_doc/hiera#transformers.HieraForImageClassification) (model HieraConfig)
- **ijepa** -- [IJepaForImageClassification](/docs/transformers/main/id/model_doc/ijepa#transformers.IJepaForImageClassification) (model IJepaConfig)
- **imagegpt** -- [ImageGPTForImageClassification](/docs/transformers/main/id/model_doc/imagegpt#transformers.ImageGPTForImageClassification) (model ImageGPTConfig)
- **levit** -- [LevitForImageClassification](/docs/transformers/main/id/model_doc/levit#transformers.LevitForImageClassification) atau [LevitForImageClassificationWithTeacher](/docs/transformers/main/id/model_doc/levit#transformers.LevitForImageClassificationWithTeacher) (model LevitConfig)
- **metaclip_2** -- [MetaClip2ForImageClassification](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2ForImageClassification) (model MetaClip2Config)
- **mobilenet_v1** -- [MobileNetV1ForImageClassification](/docs/transformers/main/id/model_doc/mobilenet_v1#transformers.MobileNetV1ForImageClassification) (model MobileNetV1Config)
- **mobilenet_v2** -- [MobileNetV2ForImageClassification](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2ForImageClassification) (model MobileNetV2Config)
- **mobilevit** -- [MobileViTForImageClassification](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTForImageClassification) (model MobileViTConfig)
- **mobilevitv2** -- [MobileViTV2ForImageClassification](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2ForImageClassification) (model MobileViTV2Config)
- **perceiver** -- [PerceiverForImageClassificationLearned](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationLearned) atau [PerceiverForImageClassificationFourier](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationFourier) atau [PerceiverForImageClassificationConvProcessing](/docs/transformers/main/id/model_doc/perceiver#transformers.PerceiverForImageClassificationConvProcessing) (model PerceiverConfig)
- **poolformer** -- [PoolFormerForImageClassification](/docs/transformers/main/id/model_doc/poolformer#transformers.PoolFormerForImageClassification) (model PoolFormerConfig)
- **pp_lcnet** -- [PPLCNetForImageClassification](/docs/transformers/main/id/model_doc/pp_lcnet#transformers.PPLCNetForImageClassification) (model PPLCNetConfig)
- **pvt** -- [PvtForImageClassification](/docs/transformers/main/id/model_doc/pvt#transformers.PvtForImageClassification) (model PvtConfig)
- **pvt_v2** -- [PvtV2ForImageClassification](/docs/transformers/main/id/model_doc/pvt_v2#transformers.PvtV2ForImageClassification) (model PvtV2Config)
- **regnet** -- [RegNetForImageClassification](/docs/transformers/main/id/model_doc/regnet#transformers.RegNetForImageClassification) (model RegNetConfig)
- **resnet** -- [ResNetForImageClassification](/docs/transformers/main/id/model_doc/resnet#transformers.ResNetForImageClassification) (model ResNetConfig)
- **segformer** -- [SegformerForImageClassification](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerForImageClassification) (model SegformerConfig)
- **shieldgemma2** -- [ShieldGemma2ForImageClassification](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2ForImageClassification) (model ShieldGemma2Config)
- **siglip** -- [SiglipForImageClassification](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipForImageClassification) (model SiglipConfig)
- **siglip2** -- [Siglip2ForImageClassification](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2ForImageClassification) (model Siglip2Config)
- **swiftformer** -- [SwiftFormerForImageClassification](/docs/transformers/main/id/model_doc/swiftformer#transformers.SwiftFormerForImageClassification) (model SwiftFormerConfig)
- **swin** -- [SwinForImageClassification](/docs/transformers/main/id/model_doc/swin#transformers.SwinForImageClassification) (model SwinConfig)
- **swinv2** -- [Swinv2ForImageClassification](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2ForImageClassification) (model Swinv2Config)
- **textnet** -- [TextNetForImageClassification](/docs/transformers/main/id/model_doc/textnet#transformers.TextNetForImageClassification) (model TextNetConfig)
- **timm_wrapper** -- [TimmWrapperForImageClassification](/docs/transformers/main/id/model_doc/timm_wrapper#transformers.TimmWrapperForImageClassification) (model TimmWrapperConfig)
- **vit** -- [ViTForImageClassification](/docs/transformers/main/id/model_doc/vit#transformers.ViTForImageClassification) (model ViTConfig)
- **vit_msn** -- [ViTMSNForImageClassification](/docs/transformers/main/id/model_doc/vit_msn#transformers.ViTMSNForImageClassification) (model ViTMSNConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForImageClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForImageClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForVideoClassification[[transformers.AutoModelForVideoClassification]]

#### transformers.AutoModelForVideoClassification[[transformers.AutoModelForVideoClassification]]

```python
transformers.AutoModelForVideoClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2514)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi video) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForVideoClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [TimesformerConfig](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerConfig): [TimesformerForVideoClassification](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerForVideoClassification) (model TimesformerConfig) - Kelas konfigurasi [VJEPA2Config](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2Config): [VJEPA2ForVideoClassification](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2ForVideoClassification) (model VJEPA2Config) - Kelas konfigurasi [VideoMAEConfig](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEConfig): [VideoMAEForVideoClassification](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEForVideoClassification) (model VideoMAEConfig) - Kelas konfigurasi [VideoPrismVisionConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismVisionConfig): [VideoPrismForVideoClassification](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismForVideoClassification) (model VideoPrismVisionConfig) - Kelas konfigurasi [VivitConfig](/docs/transformers/main/id/model_doc/vivit#transformers.VivitConfig): [VivitForVideoClassification](/docs/transformers/main/id/model_doc/vivit#transformers.VivitForVideoClassification) (model VivitConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi video) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForVideoClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForVideoClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForVideoClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi video) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **timesformer** -- [TimesformerForVideoClassification](/docs/transformers/main/id/model_doc/timesformer#transformers.TimesformerForVideoClassification) (model TimesformerConfig)
- **videomae** -- [VideoMAEForVideoClassification](/docs/transformers/main/id/model_doc/videomae#transformers.VideoMAEForVideoClassification) (model VideoMAEConfig)
- **videoprism_vision_model** -- [VideoPrismForVideoClassification](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismForVideoClassification) (model VideoPrismVisionConfig)
- **vivit** -- [VivitForVideoClassification](/docs/transformers/main/id/model_doc/vivit#transformers.VivitForVideoClassification) (model VivitConfig)
- **vjepa2** -- [VJEPA2ForVideoClassification](/docs/transformers/main/id/model_doc/vjepa2#transformers.VJEPA2ForVideoClassification) (model VJEPA2Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForVideoClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForVideoClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForVideoClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForPoseEstimation[[transformers.AutoModelForPoseEstimation]]

#### transformers.AutoModelForPoseEstimation[[transformers.AutoModelForPoseEstimation]]

```python
transformers.AutoModelForPoseEstimation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2647)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala estimasi pose) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForPoseEstimation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config): [Sapiens2ForPoseEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForPoseEstimation) (model Sapiens2Config) - Kelas konfigurasi [VitPoseConfig](/docs/transformers/main/id/model_doc/vitpose#transformers.VitPoseConfig): [VitPoseForPoseEstimation](/docs/transformers/main/id/model_doc/vitpose#transformers.VitPoseForPoseEstimation) (model VitPoseConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala estimasi pose) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPoseEstimation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForPoseEstimation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForPoseEstimation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala estimasi pose) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **sapiens2** -- [Sapiens2ForPoseEstimation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForPoseEstimation) (model Sapiens2Config)
- **vitpose** -- [VitPoseForPoseEstimation](/docs/transformers/main/id/model_doc/vitpose#transformers.VitPoseForPoseEstimation) (model VitPoseConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForPoseEstimation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForPoseEstimation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForPoseEstimation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForKeypointDetection[[transformers.AutoModelForKeypointDetection]]

#### transformers.AutoModelForKeypointDetection[[transformers.AutoModelForKeypointDetection]]

```python
transformers.AutoModelForKeypointDetection(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2281)

### AutoModelForKeypointMatching[[transformers.AutoModelForKeypointMatching]]

#### transformers.AutoModelForKeypointMatching[[transformers.AutoModelForKeypointMatching]]

```python
transformers.AutoModelForKeypointMatching(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2285)

### AutoModelForMaskedImageModeling[[transformers.AutoModelForMaskedImageModeling]]

#### transformers.AutoModelForMaskedImageModeling[[transformers.AutoModelForMaskedImageModeling]]

```python
transformers.AutoModelForMaskedImageModeling(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2610)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan gambar bermasker) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForMaskedImageModeling.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [DeiTConfig](/docs/transformers/main/id/model_doc/deit#transformers.DeiTConfig): [DeiTForMaskedImageModeling](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForMaskedImageModeling) (model DeiTConfig) - Kelas konfigurasi [FocalNetConfig](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetConfig): [FocalNetForMaskedImageModeling](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetForMaskedImageModeling) (model FocalNetConfig) - Kelas konfigurasi [SwinConfig](/docs/transformers/main/id/model_doc/swin#transformers.SwinConfig): [SwinForMaskedImageModeling](/docs/transformers/main/id/model_doc/swin#transformers.SwinForMaskedImageModeling) (model SwinConfig) - Kelas konfigurasi [Swinv2Config](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2Config): [Swinv2ForMaskedImageModeling](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2ForMaskedImageModeling) (model Swinv2Config) - Kelas konfigurasi [ViTConfig](/docs/transformers/main/id/model_doc/vit#transformers.ViTConfig): [ViTForMaskedImageModeling](/docs/transformers/main/id/model_doc/vit#transformers.ViTForMaskedImageModeling) (model ViTConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan gambar bermasker) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMaskedImageModeling

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForMaskedImageModeling.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForMaskedImageModeling.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan gambar bermasker) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **deit** -- [DeiTForMaskedImageModeling](/docs/transformers/main/id/model_doc/deit#transformers.DeiTForMaskedImageModeling) (model DeiTConfig)
- **focalnet** -- [FocalNetForMaskedImageModeling](/docs/transformers/main/id/model_doc/focalnet#transformers.FocalNetForMaskedImageModeling) (model FocalNetConfig)
- **swin** -- [SwinForMaskedImageModeling](/docs/transformers/main/id/model_doc/swin#transformers.SwinForMaskedImageModeling) (model SwinConfig)
- **swinv2** -- [Swinv2ForMaskedImageModeling](/docs/transformers/main/id/model_doc/swinv2#transformers.Swinv2ForMaskedImageModeling) (model Swinv2Config)
- **vit** -- [ViTForMaskedImageModeling](/docs/transformers/main/id/model_doc/vit#transformers.ViTForMaskedImageModeling) (model ViTConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMaskedImageModeling

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForMaskedImageModeling.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForMaskedImageModeling.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForObjectDetection[[transformers.AutoModelForObjectDetection]]

#### transformers.AutoModelForObjectDetection[[transformers.AutoModelForObjectDetection]]

```python
transformers.AutoModelForObjectDetection(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2477)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala deteksi objek) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForObjectDetection.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [ConditionalDetrConfig](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrConfig): [ConditionalDetrForObjectDetection](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrForObjectDetection) (model ConditionalDetrConfig) - Kelas konfigurasi [DFineConfig](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineConfig): [DFineForObjectDetection](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineForObjectDetection) (model DFineConfig) - Kelas konfigurasi [DabDetrConfig](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrConfig): [DabDetrForObjectDetection](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrForObjectDetection) (model DabDetrConfig) - Kelas konfigurasi [DeformableDetrConfig](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrConfig): [DeformableDetrForObjectDetection](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrForObjectDetection) (model DeformableDetrConfig) - Kelas konfigurasi [Deimv2Config](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2Config): [Deimv2ForObjectDetection](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2ForObjectDetection) (model Deimv2Config) - Kelas konfigurasi [DetrConfig](/docs/transformers/main/id/model_doc/detr#transformers.DetrConfig): [DetrForObjectDetection](/docs/transformers/main/id/model_doc/detr#transformers.DetrForObjectDetection) (model DetrConfig) - Kelas konfigurasi [LwDetrConfig](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrConfig): [LwDetrForObjectDetection](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrForObjectDetection) (model LwDetrConfig) - Kelas konfigurasi [PPDocLayoutV2Config](/docs/transformers/main/id/model_doc/pp_doclayout_v2#transformers.PPDocLayoutV2Config): [PPDocLayoutV2ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v2#transformers.PPDocLayoutV2ForObjectDetection) (model PPDocLayoutV2Config) - Kelas konfigurasi [PPDocLayoutV3Config](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3Config): [PPDocLayoutV3ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3ForObjectDetection) (model PPDocLayoutV3Config) - Kelas konfigurasi [PPDocLayoutV4Config](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4Config): [PPDocLayoutV4ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4ForObjectDetection) (model PPDocLayoutV4Config) - Kelas konfigurasi [PPOCRV5MobileDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_det#transformers.PPOCRV5MobileDetConfig): [PPOCRV5MobileDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_det#transformers.PPOCRV5MobileDetForObjectDetection) (model PPOCRV5MobileDetConfig) - Kelas konfigurasi [PPOCRV5ServerDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv5_server_det#transformers.PPOCRV5ServerDetConfig): [PPOCRV5ServerDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv5_server_det#transformers.PPOCRV5ServerDetForObjectDetection) (model PPOCRV5ServerDetConfig) - Kelas konfigurasi [PPOCRV6MediumDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetConfig): [PPOCRV6MediumDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetForObjectDetection) (model PPOCRV6MediumDetConfig) - Kelas konfigurasi [PPOCRV6SmallDetConfig](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetConfig): [PPOCRV6SmallDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetForObjectDetection) (model PPOCRV6SmallDetConfig) - Kelas konfigurasi [RTDetrConfig](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrConfig): [RTDetrForObjectDetection](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrForObjectDetection) (model RTDetrConfig) - Kelas konfigurasi [RTDetrV2Config](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2Config): [RTDetrV2ForObjectDetection](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2ForObjectDetection) (model RTDetrV2Config) - Kelas konfigurasi [RfDetrConfig](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrConfig): [RfDetrForObjectDetection](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrForObjectDetection) (model RfDetrConfig) - Kelas konfigurasi [TableTransformerConfig](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerConfig): [TableTransformerForObjectDetection](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerForObjectDetection) (model TableTransformerConfig) - Kelas konfigurasi [YolosConfig](/docs/transformers/main/id/model_doc/yolos#transformers.YolosConfig): [YolosForObjectDetection](/docs/transformers/main/id/model_doc/yolos#transformers.YolosForObjectDetection) (model YolosConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala deteksi objek) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForObjectDetection

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForObjectDetection.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForObjectDetection.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala deteksi objek) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **conditional_detr** -- [ConditionalDetrForObjectDetection](/docs/transformers/main/id/model_doc/conditional_detr#transformers.ConditionalDetrForObjectDetection) (model ConditionalDetrConfig)
- **d_fine** -- [DFineForObjectDetection](/docs/transformers/main/id/model_doc/d_fine#transformers.DFineForObjectDetection) (model DFineConfig)
- **dab-detr** -- [DabDetrForObjectDetection](/docs/transformers/main/id/model_doc/dab-detr#transformers.DabDetrForObjectDetection) (model DabDetrConfig)
- **deformable_detr** -- [DeformableDetrForObjectDetection](/docs/transformers/main/id/model_doc/deformable_detr#transformers.DeformableDetrForObjectDetection) (model DeformableDetrConfig)
- **deimv2** -- [Deimv2ForObjectDetection](/docs/transformers/main/id/model_doc/deimv2#transformers.Deimv2ForObjectDetection) (model Deimv2Config)
- **detr** -- [DetrForObjectDetection](/docs/transformers/main/id/model_doc/detr#transformers.DetrForObjectDetection) (model DetrConfig)
- **lw_detr** -- [LwDetrForObjectDetection](/docs/transformers/main/id/model_doc/lw_detr#transformers.LwDetrForObjectDetection) (model LwDetrConfig)
- **pp_doclayout_v2** -- [PPDocLayoutV2ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v2#transformers.PPDocLayoutV2ForObjectDetection) (model PPDocLayoutV2Config)
- **pp_doclayout_v3** -- [PPDocLayoutV3ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v3#transformers.PPDocLayoutV3ForObjectDetection) (model PPDocLayoutV3Config)
- **pp_doclayout_v4** -- [PPDocLayoutV4ForObjectDetection](/docs/transformers/main/id/model_doc/pp_doclayout_v4#transformers.PPDocLayoutV4ForObjectDetection) (model PPDocLayoutV4Config)
- **pp_ocrv5_mobile_det** -- [PPOCRV5MobileDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv5_mobile_det#transformers.PPOCRV5MobileDetForObjectDetection) (model PPOCRV5MobileDetConfig)
- **pp_ocrv5_server_det** -- [PPOCRV5ServerDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv5_server_det#transformers.PPOCRV5ServerDetForObjectDetection) (model PPOCRV5ServerDetConfig)
- **pp_ocrv6_medium_det** -- [PPOCRV6MediumDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv6_medium_det#transformers.PPOCRV6MediumDetForObjectDetection) (model PPOCRV6MediumDetConfig)
- **pp_ocrv6_small_det** -- [PPOCRV6SmallDetForObjectDetection](/docs/transformers/main/id/model_doc/pp_ocrv6_small_det#transformers.PPOCRV6SmallDetForObjectDetection) (model PPOCRV6SmallDetConfig)
- **rf_detr** -- [RfDetrForObjectDetection](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrForObjectDetection) (model RfDetrConfig)
- **rt_detr** -- [RTDetrForObjectDetection](/docs/transformers/main/id/model_doc/rt_detr#transformers.RTDetrForObjectDetection) (model RTDetrConfig)
- **rt_detr_v2** -- [RTDetrV2ForObjectDetection](/docs/transformers/main/id/model_doc/rt_detr_v2#transformers.RTDetrV2ForObjectDetection) (model RTDetrV2Config)
- **table-transformer** -- [TableTransformerForObjectDetection](/docs/transformers/main/id/model_doc/table-transformer#transformers.TableTransformerForObjectDetection) (model TableTransformerConfig)
- **yolos** -- [YolosForObjectDetection](/docs/transformers/main/id/model_doc/yolos#transformers.YolosForObjectDetection) (model YolosConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForObjectDetection

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForObjectDetection.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForObjectDetection.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForImageSegmentation[[transformers.AutoModelForImageSegmentation]]

#### transformers.AutoModelForImageSegmentation[[transformers.AutoModelForImageSegmentation]]

```python
transformers.AutoModelForImageSegmentation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2434)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala segmentasi gambar) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForImageSegmentation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [DetrConfig](/docs/transformers/main/id/model_doc/detr#transformers.DetrConfig): [DetrForSegmentation](/docs/transformers/main/id/model_doc/detr#transformers.DetrForSegmentation) (model DetrConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala segmentasi gambar) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageSegmentation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForImageSegmentation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForImageSegmentation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala segmentasi gambar) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **detr** -- [DetrForSegmentation](/docs/transformers/main/id/model_doc/detr#transformers.DetrForSegmentation) (model DetrConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageSegmentation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForImageSegmentation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForImageSegmentation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForImageToImage[[transformers.AutoModelForImageToImage]]

#### transformers.AutoModelForImageToImage[[transformers.AutoModelForImageToImage]]

```python
transformers.AutoModelForImageToImage(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2293)

### AutoModelForSemanticSegmentation[[transformers.AutoModelForSemanticSegmentation]]

#### transformers.AutoModelForSemanticSegmentation[[transformers.AutoModelForSemanticSegmentation]]

```python
transformers.AutoModelForSemanticSegmentation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2441)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala segmentasi semantik) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForSemanticSegmentation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [BeitConfig](/docs/transformers/main/id/model_doc/beit#transformers.BeitConfig): [BeitForSemanticSegmentation](/docs/transformers/main/id/model_doc/beit#transformers.BeitForSemanticSegmentation) (model BeitConfig) - Kelas konfigurasi [DPTConfig](/docs/transformers/main/id/model_doc/dpt#transformers.DPTConfig): [DPTForSemanticSegmentation](/docs/transformers/main/id/model_doc/dpt#transformers.DPTForSemanticSegmentation) (model DPTConfig) - Kelas konfigurasi [Data2VecVisionConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionConfig): [Data2VecVisionForSemanticSegmentation](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionForSemanticSegmentation) (model Data2VecVisionConfig) - Kelas konfigurasi [MobileNetV2Config](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2Config): [MobileNetV2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2ForSemanticSegmentation) (model MobileNetV2Config) - Kelas konfigurasi [MobileViTConfig](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTConfig): [MobileViTForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTForSemanticSegmentation) (model MobileViTConfig) - Kelas konfigurasi [MobileViTV2Config](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2Config): [MobileViTV2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2ForSemanticSegmentation) (model MobileViTV2Config) - Kelas konfigurasi [Sapiens2Config](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2Config): [Sapiens2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForSemanticSegmentation) (model Sapiens2Config) - Kelas konfigurasi [SegformerConfig](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerConfig): [SegformerForSemanticSegmentation](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerForSemanticSegmentation) (model SegformerConfig) - Kelas konfigurasi [Tipsv2DptConfig](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptConfig): [Tipsv2DptForSemanticSegmentation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForSemanticSegmentation) (model Tipsv2DptConfig) - Kelas konfigurasi [UperNetConfig](/docs/transformers/main/id/model_doc/upernet#transformers.UperNetConfig): [UperNetForSemanticSegmentation](/docs/transformers/main/id/model_doc/upernet#transformers.UperNetForSemanticSegmentation) (model UperNetConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala segmentasi semantik) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSemanticSegmentation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForSemanticSegmentation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForSemanticSegmentation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala segmentasi semantik) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **beit** -- [BeitForSemanticSegmentation](/docs/transformers/main/id/model_doc/beit#transformers.BeitForSemanticSegmentation) (model BeitConfig)
- **data2vec-vision** -- [Data2VecVisionForSemanticSegmentation](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecVisionForSemanticSegmentation) (model Data2VecVisionConfig)
- **dpt** -- [DPTForSemanticSegmentation](/docs/transformers/main/id/model_doc/dpt#transformers.DPTForSemanticSegmentation) (model DPTConfig)
- **mobilenet_v2** -- [MobileNetV2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilenet_v2#transformers.MobileNetV2ForSemanticSegmentation) (model MobileNetV2Config)
- **mobilevit** -- [MobileViTForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilevit#transformers.MobileViTForSemanticSegmentation) (model MobileViTConfig)
- **mobilevitv2** -- [MobileViTV2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/mobilevitv2#transformers.MobileViTV2ForSemanticSegmentation) (model MobileViTV2Config)
- **sapiens2** -- [Sapiens2ForSemanticSegmentation](/docs/transformers/main/id/model_doc/sapiens2#transformers.Sapiens2ForSemanticSegmentation) (model Sapiens2Config)
- **segformer** -- [SegformerForSemanticSegmentation](/docs/transformers/main/id/model_doc/segformer#transformers.SegformerForSemanticSegmentation) (model SegformerConfig)
- **tipsv2_dpt** -- [Tipsv2DptForSemanticSegmentation](/docs/transformers/main/id/model_doc/tipsv2_dpt#transformers.Tipsv2DptForSemanticSegmentation) (model Tipsv2DptConfig)
- **upernet** -- [UperNetForSemanticSegmentation](/docs/transformers/main/id/model_doc/upernet#transformers.UperNetForSemanticSegmentation) (model UperNetConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSemanticSegmentation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForSemanticSegmentation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForSemanticSegmentation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForInstanceSegmentation[[transformers.AutoModelForInstanceSegmentation]]

#### transformers.AutoModelForInstanceSegmentation[[transformers.AutoModelForInstanceSegmentation]]

```python
transformers.AutoModelForInstanceSegmentation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2468)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala segmentasi instans) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForInstanceSegmentation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [MaskFormerConfig](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerConfig): [MaskFormerForInstanceSegmentation](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerForInstanceSegmentation) (model MaskFormerConfig) - Kelas konfigurasi [RfDetrConfig](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrConfig): [RfDetrForInstanceSegmentation](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrForInstanceSegmentation) (model RfDetrConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala segmentasi instans) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForInstanceSegmentation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForInstanceSegmentation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForInstanceSegmentation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala segmentasi instans) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **maskformer** -- [MaskFormerForInstanceSegmentation](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerForInstanceSegmentation) (model MaskFormerConfig)
- **rf_detr** -- [RfDetrForInstanceSegmentation](/docs/transformers/main/id/model_doc/rf_detr#transformers.RfDetrForInstanceSegmentation) (model RfDetrConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForInstanceSegmentation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForInstanceSegmentation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForInstanceSegmentation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForUniversalSegmentation[[transformers.AutoModelForUniversalSegmentation]]

#### transformers.AutoModelForUniversalSegmentation[[transformers.AutoModelForUniversalSegmentation]]

```python
transformers.AutoModelForUniversalSegmentation(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2459)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala segmentasi gambar universal) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForUniversalSegmentation.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [DetrConfig](/docs/transformers/main/id/model_doc/detr#transformers.DetrConfig): [DetrForSegmentation](/docs/transformers/main/id/model_doc/detr#transformers.DetrForSegmentation) (model DetrConfig) - Kelas konfigurasi [EomtConfig](/docs/transformers/main/id/model_doc/eomt#transformers.EomtConfig): [EomtForUniversalSegmentation](/docs/transformers/main/id/model_doc/eomt#transformers.EomtForUniversalSegmentation) (model EomtConfig) - Kelas konfigurasi [EomtDinov3Config](/docs/transformers/main/id/model_doc/eomt_dinov3#transformers.EomtDinov3Config): [EomtDinov3ForUniversalSegmentation](/docs/transformers/main/id/model_doc/eomt_dinov3#transformers.EomtDinov3ForUniversalSegmentation) (model EomtDinov3Config) - Kelas konfigurasi [Mask2FormerConfig](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerConfig): [Mask2FormerForUniversalSegmentation](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerForUniversalSegmentation) (model Mask2FormerConfig) - Kelas konfigurasi [MaskFormerConfig](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerConfig): [MaskFormerForInstanceSegmentation](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerForInstanceSegmentation) (model MaskFormerConfig) - Kelas konfigurasi [OneFormerConfig](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerConfig): [OneFormerForUniversalSegmentation](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerForUniversalSegmentation) (model OneFormerConfig) - Kelas konfigurasi [VideomtConfig](/docs/transformers/main/id/model_doc/videomt#transformers.VideomtConfig): [VideomtForUniversalSegmentation](/docs/transformers/main/id/model_doc/videomt#transformers.VideomtForUniversalSegmentation) (model VideomtConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala segmentasi gambar universal) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForUniversalSegmentation

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForUniversalSegmentation.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForUniversalSegmentation.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala segmentasi gambar universal) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **detr** -- [DetrForSegmentation](/docs/transformers/main/id/model_doc/detr#transformers.DetrForSegmentation) (model DetrConfig)
- **eomt** -- [EomtForUniversalSegmentation](/docs/transformers/main/id/model_doc/eomt#transformers.EomtForUniversalSegmentation) (model EomtConfig)
- **eomt_dinov3** -- [EomtDinov3ForUniversalSegmentation](/docs/transformers/main/id/model_doc/eomt_dinov3#transformers.EomtDinov3ForUniversalSegmentation) (model EomtDinov3Config)
- **mask2former** -- [Mask2FormerForUniversalSegmentation](/docs/transformers/main/id/model_doc/mask2former#transformers.Mask2FormerForUniversalSegmentation) (model Mask2FormerConfig)
- **maskformer** -- [MaskFormerForInstanceSegmentation](/docs/transformers/main/id/model_doc/maskformer#transformers.MaskFormerForInstanceSegmentation) (model MaskFormerConfig)
- **oneformer** -- [OneFormerForUniversalSegmentation](/docs/transformers/main/id/model_doc/oneformer#transformers.OneFormerForUniversalSegmentation) (model OneFormerConfig)
- **videomt** -- [VideomtForUniversalSegmentation](/docs/transformers/main/id/model_doc/videomt#transformers.VideomtForUniversalSegmentation) (model VideomtConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForUniversalSegmentation

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForUniversalSegmentation.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForUniversalSegmentation.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForZeroShotImageClassification[[transformers.AutoModelForZeroShotImageClassification]]

#### transformers.AutoModelForZeroShotImageClassification[[transformers.AutoModelForZeroShotImageClassification]]

```python
transformers.AutoModelForZeroShotImageClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2425)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar nol-tembakan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForZeroShotImageClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [AlignConfig](/docs/transformers/main/id/model_doc/align#transformers.AlignConfig): [AlignModel](/docs/transformers/main/id/model_doc/align#transformers.AlignModel) (model AlignConfig) - Kelas konfigurasi [AltCLIPConfig](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPConfig): [AltCLIPModel](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPModel) (model AltCLIPConfig) - Kelas konfigurasi [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config): [Blip2ForImageTextRetrieval](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForImageTextRetrieval) (model Blip2Config) - Kelas konfigurasi [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig): [BlipModel](/docs/transformers/main/id/model_doc/blip#transformers.BlipModel) (model BlipConfig) - Kelas konfigurasi [CLIPConfig](/docs/transformers/main/id/model_doc/clip#transformers.CLIPConfig): [CLIPModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPModel) (model CLIPConfig) - Kelas konfigurasi [CLIPSegConfig](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegConfig): [CLIPSegModel](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegModel) (model CLIPSegConfig) - Kelas konfigurasi [ChineseCLIPConfig](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPConfig): [ChineseCLIPModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPModel) (model ChineseCLIPConfig) - Kelas konfigurasi [MetaClip2Config](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Config): [MetaClip2Model](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Model) (model MetaClip2Config) - Kelas konfigurasi [Siglip2Config](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Config): [Siglip2Model](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Model) (model Siglip2Config) - Kelas konfigurasi [SiglipConfig](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipConfig): [SiglipModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipModel) (model SiglipConfig) - Kelas konfigurasi [Tipsv2Config](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Config): [Tipsv2Model](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Model) (model Tipsv2Config) - Kelas konfigurasi [VideoPrismConfig](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismConfig): [VideoPrismClipModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismClipModel) (model VideoPrismConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar nol-shot) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForZeroShotImageClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForZeroShotImageClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForZeroShotImageClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi gambar nol-shot) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **align** -- [AlignModel](/docs/transformers/main/id/model_doc/align#transformers.AlignModel) (model AlignConfig)
- **altclip** -- [AltCLIPModel](/docs/transformers/main/id/model_doc/altclip#transformers.AltCLIPModel) (model AltCLIPConfig)
- **blip** -- [BlipModel](/docs/transformers/main/id/model_doc/blip#transformers.BlipModel) (model BlipConfig)
- **blip-2** -- [Blip2ForImageTextRetrieval](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForImageTextRetrieval) (model Blip2Config)
- **chinese_clip** -- [ChineseCLIPModel](/docs/transformers/main/id/model_doc/chinese_clip#transformers.ChineseCLIPModel) (model ChineseCLIPConfig)
- **clip** -- [CLIPModel](/docs/transformers/main/id/model_doc/clip#transformers.CLIPModel) (model CLIPConfig)
- **clipseg** -- [CLIPSegModel](/docs/transformers/main/id/model_doc/clipseg#transformers.CLIPSegModel) (model CLIPSegConfig)
- **metaclip_2** -- [MetaClip2Model](/docs/transformers/main/id/model_doc/metaclip_2#transformers.MetaClip2Model) (model MetaClip2Config)
- **siglip** -- [SiglipModel](/docs/transformers/main/id/model_doc/siglip#transformers.SiglipModel) (model SiglipConfig)
- **siglip2** -- [Siglip2Model](/docs/transformers/main/id/model_doc/siglip2#transformers.Siglip2Model) (model Siglip2Config)
- **tipsv2** -- [Tipsv2Model](/docs/transformers/main/id/model_doc/tipsv2#transformers.Tipsv2Model) (model Tipsv2Config)
- **videoprism** -- [VideoPrismClipModel](/docs/transformers/main/id/model_doc/videoprism#transformers.VideoPrismClipModel) (model VideoPrismConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForZeroShotImageClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForZeroShotImageClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForZeroShotImageClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForZeroShotObjectDetection[[transformers.AutoModelForZeroShotObjectDetection]]

#### transformers.AutoModelForZeroShotObjectDetection[[transformers.AutoModelForZeroShotObjectDetection]]

```python
transformers.AutoModelForZeroShotObjectDetection(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2484)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala deteksi objek zero-shot) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForZeroShotObjectDetection.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [GroundingDinoConfig](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoConfig): [GroundingDinoForObjectDetection](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoForObjectDetection) (model GroundingDinoConfig) - Kelas konfigurasi [MMGroundingDinoConfig](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoConfig): [MMGroundingDinoForObjectDetection](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoForObjectDetection) (model MMGroundingDinoConfig) - Kelas konfigurasi [OmDetTurboConfig](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboConfig): [OmDetTurboForObjectDetection](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboForObjectDetection) (model OmDetTurboConfig) - Kelas konfigurasi [OwlViTConfig](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTConfig): [OwlViTForObjectDetection](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTForObjectDetection) (model OwlViTConfig) - Kelas konfigurasi [Owlv2Config](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2Config): [Owlv2ForObjectDetection](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2ForObjectDetection) (model Owlv2Config)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala deteksi objek zero-shot) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForZeroShotObjectDetection

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForZeroShotObjectDetection.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForZeroShotObjectDetection.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala deteksi objek zero-shot) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **grounding-dino** -- [GroundingDinoForObjectDetection](/docs/transformers/main/id/model_doc/grounding-dino#transformers.GroundingDinoForObjectDetection) (model GroundingDinoConfig)
- **mm-grounding-dino** -- [MMGroundingDinoForObjectDetection](/docs/transformers/main/id/model_doc/mm-grounding-dino#transformers.MMGroundingDinoForObjectDetection) (model MMGroundingDinoConfig)
- **omdet-turbo** -- [OmDetTurboForObjectDetection](/docs/transformers/main/id/model_doc/omdet-turbo#transformers.OmDetTurboForObjectDetection) (model OmDetTurboConfig)
- **owlv2** -- [Owlv2ForObjectDetection](/docs/transformers/main/id/model_doc/owlv2#transformers.Owlv2ForObjectDetection) (model Owlv2Config)
- **owlvit** -- [OwlViTForObjectDetection](/docs/transformers/main/id/model_doc/owlvit#transformers.OwlViTForObjectDetection) (model OwlViTConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForZeroShotObjectDetection

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForZeroShotObjectDetection.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForZeroShotObjectDetection.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

## Audio

Kelas auto berikut tersedia untuk tugas audio berikut.

### AutoModelForAudioClassification[[transformers.AutoModelForAudioClassification]]

#### transformers.AutoModelForAudioClassification[[transformers.AutoModelForAudioClassification]]

```python
transformers.AutoModelForAudioClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2545)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi audio) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForAudioClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [ASTConfig](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTConfig): [ASTForAudioClassification](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTForAudioClassification) (model ASTConfig) - Kelas konfigurasi [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig): [Data2VecAudioForSequenceClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForSequenceClassification) (model Data2VecAudioConfig) - Kelas konfigurasi [HubertConfig](/docs/transformers/main/id/model_doc/hubert#transformers.HubertConfig): [HubertForSequenceClassification](/docs/transformers/main/id/model_doc/hubert#transformers.HubertForSequenceClassification) (model HubertConfig) - Kelas konfigurasi [SEWConfig](/docs/transformers/main/id/model_doc/sew#transformers.SEWConfig): [SEWForSequenceClassification](/docs/transformers/main/id/model_doc/sew#transformers.SEWForSequenceClassification) (model SEWConfig) - Kelas konfigurasi [SEWDConfig](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDConfig): [SEWDForSequenceClassification](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDForSequenceClassification) (model SEWDConfig) - Kelas konfigurasi [UniSpeechConfig](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechConfig): [UniSpeechForSequenceClassification](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForSequenceClassification) (model UniSpeechConfig) - Kelas konfigurasi [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig): [UniSpeechSatForSequenceClassification](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForSequenceClassification) (model UniSpeechSatConfig) - Kelas konfigurasi [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig): [Wav2Vec2BertForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForSequenceClassification) (model Wav2Vec2BertConfig) - Kelas konfigurasi [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config): [Wav2Vec2ForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForSequenceClassification) (model Wav2Vec2Config) - Kelas konfigurasi [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig): [Wav2Vec2ConformerForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForSequenceClassification) (model Wav2Vec2ConformerConfig) - Kelas konfigurasi [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig): [WavLMForSequenceClassification](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForSequenceClassification) (model WavLMConfig) - Kelas konfigurasi [WhisperConfig](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperConfig): [WhisperForAudioClassification](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForAudioClassification) (model WhisperConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi audio) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForAudioClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForAudioClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi audio) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **audio-spectrogram-transformer** -- [ASTForAudioClassification](/docs/transformers/main/id/model_doc/audio-spectrogram-transformer#transformers.ASTForAudioClassification) (model ASTConfig)
- **data2vec-audio** -- [Data2VecAudioForSequenceClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForSequenceClassification) (model Data2VecAudioConfig)
- **hubert** -- [HubertForSequenceClassification](/docs/transformers/main/id/model_doc/hubert#transformers.HubertForSequenceClassification) (model HubertConfig)
- **sew** -- [SEWForSequenceClassification](/docs/transformers/main/id/model_doc/sew#transformers.SEWForSequenceClassification) (model SEWConfig)
- **sew-d** -- [SEWDForSequenceClassification](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDForSequenceClassification) (model SEWDConfig)
- **unispeech** -- [UniSpeechForSequenceClassification](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForSequenceClassification) (model UniSpeechConfig)
- **unispeech-sat** -- [UniSpeechSatForSequenceClassification](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForSequenceClassification) (model UniSpeechSatConfig)
- **wav2vec2** -- [Wav2Vec2ForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForSequenceClassification) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForSequenceClassification) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerForSequenceClassification](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForSequenceClassification) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMForSequenceClassification](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForSequenceClassification) (model WavLMConfig)
- **whisper** -- [WhisperForAudioClassification](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForAudioClassification) (model WhisperConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForAudioClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForAudioClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForAudioFrameClassification[[transformers.AutoModelForAudioFrameClassification]]

#### transformers.AutoModelForAudioFrameClassification[[transformers.AutoModelForAudioFrameClassification]]

```python
transformers.AutoModelForAudioFrameClassification(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2582)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi frame (token) audio) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForAudioFrameClassification.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig): [Data2VecAudioForAudioFrameClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForAudioFrameClassification) (model Data2VecAudioConfig) - Kelas konfigurasi [Nemotron3DiarizationConfig](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationConfig): [Nemotron3DiarizationForAudioFrameClassification](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationForAudioFrameClassification) (model Nemotron3DiarizationConfig) - Kelas konfigurasi [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig): [UniSpeechSatForAudioFrameClassification](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForAudioFrameClassification) (model UniSpeechSatConfig) - Kelas konfigurasi [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig): [Wav2Vec2BertForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForAudioFrameClassification) (model Wav2Vec2BertConfig) - Kelas konfigurasi [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config): [Wav2Vec2ForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForAudioFrameClassification) (model Wav2Vec2Config) - Kelas konfigurasi [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig): [Wav2Vec2ConformerForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForAudioFrameClassification) (model Wav2Vec2ConformerConfig) - Kelas konfigurasi [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig): [WavLMForAudioFrameClassification](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForAudioFrameClassification) (model WavLMConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi frame (token) audio) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioFrameClassification

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForAudioFrameClassification.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForAudioFrameClassification.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi frame (token) audio) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **data2vec-audio** -- [Data2VecAudioForAudioFrameClassification](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForAudioFrameClassification) (model Data2VecAudioConfig)
- **nemotron3_diarization** -- [Nemotron3DiarizationForAudioFrameClassification](/docs/transformers/main/id/model_doc/nemotron3_diarization#transformers.Nemotron3DiarizationForAudioFrameClassification) (model Nemotron3DiarizationConfig)
- **unispeech-sat** -- [UniSpeechSatForAudioFrameClassification](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForAudioFrameClassification) (model UniSpeechSatConfig)
- **wav2vec2** -- [Wav2Vec2ForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForAudioFrameClassification) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForAudioFrameClassification) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerForAudioFrameClassification](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForAudioFrameClassification) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMForAudioFrameClassification](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForAudioFrameClassification) (model WavLMConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioFrameClassification

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForAudioFrameClassification.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForAudioFrameClassification.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForCTC[[transformers.AutoModelForCTC]]

#### transformers.AutoModelForCTC[[transformers.AutoModelForCTC]]

```python
transformers.AutoModelForCTC(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2552)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala klasifikasi temporal koneksionis) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForCTC.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig): [Data2VecAudioForCTC](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForCTC) (model Data2VecAudioConfig) - Kelas konfigurasi [GraniteSpeech5CTCConfig](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5CTCConfig): [GraniteSpeech5ForCTC](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5ForCTC) (model GraniteSpeech5CTCConfig) - Kelas konfigurasi [HubertConfig](/docs/transformers/main/id/model_doc/hubert#transformers.HubertConfig): [HubertForCTC](/docs/transformers/main/id/model_doc/hubert#transformers.HubertForCTC) (model HubertConfig) - Kelas konfigurasi [LasrCTCConfig](/docs/transformers/main/id/model_doc/lasr#transformers.LasrCTCConfig): [LasrForCTC](/docs/transformers/main/id/model_doc/lasr#transformers.LasrForCTC) (model LasrCTCConfig) - Kelas konfigurasi [ParakeetCTCConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetCTCConfig): [ParakeetForCTC](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForCTC) (model ParakeetCTCConfig) - Kelas konfigurasi [SEWConfig](/docs/transformers/main/id/model_doc/sew#transformers.SEWConfig): [SEWForCTC](/docs/transformers/main/id/model_doc/sew#transformers.SEWForCTC) (model SEWConfig) - Kelas konfigurasi [SEWDConfig](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDConfig): [SEWDForCTC](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDForCTC) (model SEWDConfig) - Kelas konfigurasi [UniSpeechConfig](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechConfig): [UniSpeechForCTC](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForCTC) (model UniSpeechConfig) - Kelas konfigurasi [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig): [UniSpeechSatForCTC](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForCTC) (model UniSpeechSatConfig) - Kelas konfigurasi [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig): [Wav2Vec2BertForCTC](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForCTC) (model Wav2Vec2BertConfig) - Kelas konfigurasi [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config): [Wav2Vec2ForCTC](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForCTC) (model Wav2Vec2Config) - Kelas konfigurasi [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig): [Wav2Vec2ConformerForCTC](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForCTC) (model Wav2Vec2ConformerConfig) - Kelas konfigurasi [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig): [WavLMForCTC](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForCTC) (model WavLMConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala klasifikasi temporal koneksionis) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForCTC

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForCTC.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForCTC.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala klasifikasi temporal koneksionis) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **data2vec-audio** -- [Data2VecAudioForCTC](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForCTC) (model Data2VecAudioConfig)
- **granite_speech5_ctc** -- [GraniteSpeech5ForCTC](/docs/transformers/main/id/model_doc/granite_speech5#transformers.GraniteSpeech5ForCTC) (model GraniteSpeech5CTCConfig)
- **hubert** -- [HubertForCTC](/docs/transformers/main/id/model_doc/hubert#transformers.HubertForCTC) (model HubertConfig)
- **lasr_ctc** -- [LasrForCTC](/docs/transformers/main/id/model_doc/lasr#transformers.LasrForCTC) (model LasrCTCConfig)
- **parakeet_ctc** -- [ParakeetForCTC](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForCTC) (model ParakeetCTCConfig)
- **sew** -- [SEWForCTC](/docs/transformers/main/id/model_doc/sew#transformers.SEWForCTC) (model SEWConfig)
- **sew-d** -- [SEWDForCTC](/docs/transformers/main/id/model_doc/sew-d#transformers.SEWDForCTC) (model SEWDConfig)
- **unispeech** -- [UniSpeechForCTC](/docs/transformers/main/id/model_doc/unispeech#transformers.UniSpeechForCTC) (model UniSpeechConfig)
- **unispeech-sat** -- [UniSpeechSatForCTC](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForCTC) (model UniSpeechSatConfig)
- **wav2vec2** -- [Wav2Vec2ForCTC](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForCTC) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertForCTC](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForCTC) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerForCTC](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForCTC) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMForCTC](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForCTC) (model WavLMConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForCTC

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForCTC.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForCTC.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTDT[[transformers.AutoModelForTDT]]

#### transformers.AutoModelForTDT[[transformers.AutoModelForTDT]]

```python
transformers.AutoModelForTDT(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2566)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala transducer token-dan-durasi) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTDT.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [ParakeetTDTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetTDTConfig): [ParakeetForTDT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForTDT) (model ParakeetTDTConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala transduser token-dan-durasi) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTDT

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForTDT.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTDT.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala transducer token-dan-durasi) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **parakeet_tdt** -- [ParakeetForTDT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForTDT) (model ParakeetTDTConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTDT

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTDT.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForTDT.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForRNNT[[transformers.AutoModelForRNNT]]

#### transformers.AutoModelForRNNT[[transformers.AutoModelForRNNT]]

```python
transformers.AutoModelForRNNT(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2559)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala transducer RNN) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForRNNT.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Nemotron3_5AsrConfig](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrConfig): [Nemotron3_5AsrForRNNT](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrForRNNT) (model Nemotron3_5AsrConfig) - Kelas konfigurasi [NemotronAsrStreamingConfig](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingConfig): [NemotronAsrStreamingForRNNT](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingForRNNT) (model NemotronAsrStreamingConfig) - Kelas konfigurasi [ParakeetRNNTConfig](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetRNNTConfig): [ParakeetForRNNT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForRNNT) (model ParakeetRNNTConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala transducer RNN) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForRNNT

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForRNNT.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForRNNT.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala transducer RNN) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **nemotron3_5_asr** -- [Nemotron3_5AsrForRNNT](/docs/transformers/main/id/model_doc/nemotron3_5_asr#transformers.Nemotron3_5AsrForRNNT) (model Nemotron3_5AsrConfig)
- **nemotron_asr_streaming** -- [NemotronAsrStreamingForRNNT](/docs/transformers/main/id/model_doc/nemotron_asr_streaming#transformers.NemotronAsrStreamingForRNNT) (model NemotronAsrStreamingConfig)
- **parakeet_rnnt** -- [ParakeetForRNNT](/docs/transformers/main/id/model_doc/parakeet#transformers.ParakeetForRNNT) (model ParakeetRNNTConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForRNNT

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForRNNT.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForRNNT.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForSpeechSeq2Seq[[transformers.AutoModelForSpeechSeq2Seq]]

#### transformers.AutoModelForSpeechSeq2Seq[[transformers.AutoModelForSpeechSeq2Seq]]

```python
transformers.AutoModelForSpeechSeq2Seq(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2573)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan ucapan-ke-teks berurutan-ke-urutan) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForSpeechSeq2Seq.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [CanaryConfig](/docs/transformers/main/id/model_doc/canary#transformers.CanaryConfig): [CanaryForConditionalGeneration](/docs/transformers/main/id/model_doc/canary#transformers.CanaryForConditionalGeneration) (model CanaryConfig) - Kelas konfigurasi [CohereAsrConfig](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrConfig): [CohereAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrForConditionalGeneration) (model CohereAsrConfig) - Kelas konfigurasi [DiaConfig](/docs/transformers/main/id/model_doc/dia#transformers.DiaConfig): [DiaForConditionalGeneration](/docs/transformers/main/id/model_doc/dia#transformers.DiaForConditionalGeneration) (model DiaConfig) - Kelas konfigurasi [FunAsrNanoConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoConfig): [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (model FunAsrNanoConfig) - Kelas konfigurasi [GraniteSpeechConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechConfig): [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (model GraniteSpeechConfig) - Kelas konfigurasi [GraniteSpeechPlusConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusConfig): [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (model GraniteSpeechPlusConfig) - Kelas konfigurasi [KyutaiSpeechToTextConfig](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextConfig): [KyutaiSpeechToTextForConditionalGeneration](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextForConditionalGeneration) (model KyutaiSpeechToTextConfig) - Kelas konfigurasi [MoonshineConfig](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineConfig): [MoonshineForConditionalGeneration](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineForConditionalGeneration) (model MoonshineConfig) - Kelas konfigurasi [MoonshineStreamingConfig](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingConfig): [MoonshineStreamingForConditionalGeneration](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingForConditionalGeneration) (model MoonshineStreamingConfig) - Kelas konfigurasi [Pop2PianoConfig](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoConfig): [Pop2PianoForConditionalGeneration](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoForConditionalGeneration) (model Pop2PianoConfig) - Kelas konfigurasi [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig): [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig) - Kelas konfigurasi [SeamlessM4TConfig](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TConfig): [SeamlessM4TForSpeechToText](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TForSpeechToText) (model SeamlessM4TConfig) - Kelas konfigurasi [SeamlessM4Tv2Config](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2Config): [SeamlessM4Tv2ForSpeechToText](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2ForSpeechToText) (model SeamlessM4Tv2Config) - Kelas konfigurasi [Speech2TextConfig](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextConfig): [Speech2TextForConditionalGeneration](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextForConditionalGeneration) (model Speech2TextConfig) - Kelas konfigurasi [SpeechEncoderDecoderConfig](/docs/transformers/main/id/model_doc/speech-encoder-decoder#transformers.SpeechEncoderDecoderConfig): [SpeechEncoderDecoderModel](/docs/transformers/main/id/model_doc/speech-encoder-decoder#transformers.SpeechEncoderDecoderModel) (model SpeechEncoderDecoderConfig) - Kelas konfigurasi [SpeechT5Config](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5Config): [SpeechT5ForSpeechToText](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5ForSpeechToText) (model SpeechT5Config) - Kelas konfigurasi [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig): [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig) - Kelas konfigurasi [VibeVoiceConfig](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceConfig): [VibeVoiceForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceForConditionalGeneration) (model VibeVoiceConfig) - Kelas konfigurasi [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig): [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig) - Kelas konfigurasi [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig): [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig) - Kelas konfigurasi [WhisperConfig](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperConfig): [WhisperForConditionalGeneration](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForConditionalGeneration) (model WhisperConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan ucapan-ke-teks berurutan-ke-urutan) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSpeechSeq2Seq

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForSpeechSeq2Seq.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForSpeechSeq2Seq.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan ucapan-ke-teks berurutan-ke-urutan) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **canary** -- [CanaryForConditionalGeneration](/docs/transformers/main/id/model_doc/canary#transformers.CanaryForConditionalGeneration) (model CanaryConfig)
- **cohere_asr** -- [CohereAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_asr#transformers.CohereAsrForConditionalGeneration) (model CohereAsrConfig)
- **dia** -- [DiaForConditionalGeneration](/docs/transformers/main/id/model_doc/dia#transformers.DiaForConditionalGeneration) (model DiaConfig)
- **fun_asr_nano** -- [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (model FunAsrNanoConfig)
- **granite_speech** -- [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (model GraniteSpeechConfig)
- **granite_speech_plus** -- [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (model GraniteSpeechPlusConfig)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextForConditionalGeneration](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextForConditionalGeneration) (model KyutaiSpeechToTextConfig)
- **moonshine** -- [MoonshineForConditionalGeneration](/docs/transformers/main/id/model_doc/moonshine#transformers.MoonshineForConditionalGeneration) (model MoonshineConfig)
- **moonshine_streaming** -- [MoonshineStreamingForConditionalGeneration](/docs/transformers/main/id/model_doc/moonshine_streaming#transformers.MoonshineStreamingForConditionalGeneration) (model MoonshineStreamingConfig)
- **pop2piano** -- [Pop2PianoForConditionalGeneration](/docs/transformers/main/id/model_doc/pop2piano#transformers.Pop2PianoForConditionalGeneration) (model Pop2PianoConfig)
- **qwen3_asr** -- [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig)
- **seamless_m4t** -- [SeamlessM4TForSpeechToText](/docs/transformers/main/id/model_doc/seamless_m4t#transformers.SeamlessM4TForSpeechToText) (model SeamlessM4TConfig)
- **seamless_m4t_v2** -- [SeamlessM4Tv2ForSpeechToText](/docs/transformers/main/id/model_doc/seamless_m4t_v2#transformers.SeamlessM4Tv2ForSpeechToText) (model SeamlessM4Tv2Config)
- **speech-encoder-decoder** -- [SpeechEncoderDecoderModel](/docs/transformers/main/id/model_doc/speech-encoder-decoder#transformers.SpeechEncoderDecoderModel) (model SpeechEncoderDecoderConfig)
- **speech_to_text** -- [Speech2TextForConditionalGeneration](/docs/transformers/main/id/model_doc/speech_to_text#transformers.Speech2TextForConditionalGeneration) (model Speech2TextConfig)
- **speecht5** -- [SpeechT5ForSpeechToText](/docs/transformers/main/id/model_doc/speecht5#transformers.SpeechT5ForSpeechToText) (model SpeechT5Config)
- **vibevoice** -- [VibeVoiceForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice#transformers.VibeVoiceForConditionalGeneration) (model VibeVoiceConfig)
- **vibevoice_asr** -- [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig)
- **voxtral** -- [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig)
- **whisper** -- [WhisperForConditionalGeneration](/docs/transformers/main/id/model_doc/whisper#transformers.WhisperForConditionalGeneration) (model WhisperConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForSpeechSeq2Seq

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForSpeechSeq2Seq.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForSpeechSeq2Seq.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForAudioXVector[[transformers.AutoModelForAudioXVector]]

#### transformers.AutoModelForAudioXVector[[transformers.AutoModelForAudioXVector]]

```python
transformers.AutoModelForAudioXVector(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2591)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan audio retrieval melalui x-vector head) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForAudioXVector.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [Data2VecAudioConfig](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioConfig): [Data2VecAudioForXVector](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForXVector) (model Data2VecAudioConfig) - Kelas konfigurasi [UniSpeechSatConfig](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatConfig): [UniSpeechSatForXVector](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForXVector) (model UniSpeechSatConfig) - Kelas konfigurasi [Wav2Vec2BertConfig](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertConfig): [Wav2Vec2BertForXVector](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForXVector) (model Wav2Vec2BertConfig) - Kelas konfigurasi [Wav2Vec2Config](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2Config): [Wav2Vec2ForXVector](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForXVector) (model Wav2Vec2Config) - Kelas konfigurasi [Wav2Vec2ConformerConfig](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerConfig): [Wav2Vec2ConformerForXVector](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForXVector) (model Wav2Vec2ConformerConfig) - Kelas konfigurasi [WavLMConfig](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMConfig): [WavLMForXVector](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForXVector) (model WavLMConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan audio retrieval melalui kepala x-vector) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioXVector

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForAudioXVector.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForAudioXVector.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan audio retrieval melalui x-vector head) dari model pretrained.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **data2vec-audio** -- [Data2VecAudioForXVector](/docs/transformers/main/id/model_doc/data2vec#transformers.Data2VecAudioForXVector) (model Data2VecAudioConfig)
- **unispeech-sat** -- [UniSpeechSatForXVector](/docs/transformers/main/id/model_doc/unispeech-sat#transformers.UniSpeechSatForXVector) (model UniSpeechSatConfig)
- **wav2vec2** -- [Wav2Vec2ForXVector](/docs/transformers/main/id/model_doc/wav2vec2#transformers.Wav2Vec2ForXVector) (model Wav2Vec2Config)
- **wav2vec2-bert** -- [Wav2Vec2BertForXVector](/docs/transformers/main/id/model_doc/wav2vec2-bert#transformers.Wav2Vec2BertForXVector) (model Wav2Vec2BertConfig)
- **wav2vec2-conformer** -- [Wav2Vec2ConformerForXVector](/docs/transformers/main/id/model_doc/wav2vec2-conformer#transformers.Wav2Vec2ConformerForXVector) (model Wav2Vec2ConformerConfig)
- **wavlm** -- [WavLMForXVector](/docs/transformers/main/id/model_doc/wavlm#transformers.WavLMForXVector) (model WavLMConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioXVector

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForAudioXVector.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForAudioXVector.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTextToSpectrogram[[transformers.AutoModelForTextToSpectrogram]]

#### transformers.AutoModelForTextToSpectrogram[[transformers.AutoModelForTextToSpectrogram]]

```python
transformers.AutoModelForTextToSpectrogram(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2595)

### AutoModelForTextToWaveform[[transformers.AutoModelForTextToWaveform]]

#### transformers.AutoModelForTextToWaveform[[transformers.AutoModelForTextToWaveform]]

```python
transformers.AutoModelForTextToWaveform(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2599)

### AutoModelForAudioTokenization[[transformers.AutoModelForAudioTokenization]]

#### transformers.AutoModelForAudioTokenization[[transformers.AutoModelForAudioTokenization]]

```python
transformers.AutoModelForAudioTokenization(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2617)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan tokenisasi audio melalui kepala codebooks) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForAudioTokenization.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [DacConfig](/docs/transformers/main/id/model_doc/dac#transformers.DacConfig): [DacModel](/docs/transformers/main/id/model_doc/dac#transformers.DacModel) (model DacConfig) - Kelas konfigurasi [HiggsAudioV2TokenizerConfig](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerConfig): [HiggsAudioV2TokenizerModel](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerModel) (model HiggsAudioV2TokenizerConfig) - Kelas konfigurasi [NeuCodecConfig](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecConfig): [NeuCodecModel](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecModel) (model NeuCodecConfig) - Kelas konfigurasi [PocketTTSAudioTokenizerConfig](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerConfig): [PocketTTSAudioTokenizerModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerModel) (model PocketTTSAudioTokenizerConfig) - Kelas konfigurasi [VibeVoiceAcousticTokenizerConfig](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerConfig): [VibeVoiceAcousticTokenizerModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerModel) (model VibeVoiceAcousticTokenizerConfig) - Kelas konfigurasi [Xcodec2Config](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Config): [Xcodec2Model](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Model) (model Xcodec2Config)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan tokenisasi audio melalui kepala codebook) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioTokenization

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForAudioTokenization.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForAudioTokenization.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan tokenisasi audio melalui kepala codebook) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **dac** -- [DacModel](/docs/transformers/main/id/model_doc/dac#transformers.DacModel) (model DacConfig)
- **higgs_audio_v2_tokenizer** -- [HiggsAudioV2TokenizerModel](/docs/transformers/main/id/model_doc/higgs_audio_v2_tokenizer#transformers.HiggsAudioV2TokenizerModel) (model HiggsAudioV2TokenizerConfig)
- **neucodec** -- [NeuCodecModel](/docs/transformers/main/id/model_doc/neucodec#transformers.NeuCodecModel) (model NeuCodecConfig)
- **pocket_tts_audio_tokenizer** -- [PocketTTSAudioTokenizerModel](/docs/transformers/main/id/model_doc/pocket_tts_audio_tokenizer#transformers.PocketTTSAudioTokenizerModel) (model PocketTTSAudioTokenizerConfig)
- **vibevoice_acoustic_tokenizer** -- [VibeVoiceAcousticTokenizerModel](/docs/transformers/main/id/model_doc/vibevoice_acoustic_tokenizer#transformers.VibeVoiceAcousticTokenizerModel) (model VibeVoiceAcousticTokenizerConfig)
- **xcodec2** -- [Xcodec2Model](/docs/transformers/main/id/model_doc/xcodec2#transformers.Xcodec2Model) (model Xcodec2Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForAudioTokenization

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForAudioTokenization.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForAudioTokenization.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

## Multimodal

Kelas auto berikut tersedia untuk tugas multimodal berikut.

### AutoModelForMultimodalLM[[transformers.AutoModelForMultimodalLM]]

#### transformers.AutoModelForMultimodalLM[[transformers.AutoModelForMultimodalLM]]

```python
transformers.AutoModelForMultimodalLM(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2538)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala generasi multimodal) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForMultimodalLM.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AriaConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaConfig) configuration class: [AriaForConditionalGeneration](/docs/transformers/main/id/model_doc/aria#transformers.AriaForConditionalGeneration) (AriaConfig model) - [AyaVisionConfig](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionConfig) configuration class: [AyaVisionForConditionalGeneration](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionForConditionalGeneration) (AyaVisionConfig model) - [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config) configuration class: [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (Blip2Config model) - [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig) configuration class: [BlipForConditionalGeneration](/docs/transformers/main/id/model_doc/blip#transformers.BlipForConditionalGeneration) (BlipConfig model) - [ChameleonConfig](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonConfig) configuration class: [ChameleonForConditionalGeneration](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonForConditionalGeneration) (ChameleonConfig model) - [Cohere2VisionConfig](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionConfig) configuration class: [Cohere2VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionForConditionalGeneration) (Cohere2VisionConfig model) - [CohereCompassConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassConfig) configuration class: [CohereCompassForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForConditionalGeneration) (CohereCompassConfig model) - [Cosmos3EdgeConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeConfig) configuration class: [Cosmos3EdgeForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeForConditionalGeneration) (Cosmos3EdgeConfig model) - [Cosmos3OmniConfig](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniConfig) configuration class: [Cosmos3OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniForConditionalGeneration) (Cosmos3OmniConfig model) - [DeepseekOcr2Config](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Config) configuration class: [DeepseekOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2ForConditionalGeneration) (DeepseekOcr2Config model) - [DeepseekVLConfig](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLConfig) configuration class: [DeepseekVLForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLForConditionalGeneration) (DeepseekVLConfig model) - [DeepseekVLHybridConfig](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridConfig) configuration class: [DeepseekVLHybridForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridForConditionalGeneration) (DeepseekVLHybridConfig model) - [DiffusionGemmaConfig](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaConfig) configuration class: [DiffusionGemmaForBlockDiffusion](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaForBlockDiffusion) (DiffusionGemmaConfig model) - [Emu3Config](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Config) configuration class: [Emu3ForConditionalGeneration](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForConditionalGeneration) (Emu3Config model) - [Ernie4_5_VLMoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeConfig) configuration class: [Ernie4_5_VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeForConditionalGeneration) (Ernie4_5_VLMoeConfig model) - [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) configuration class: [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (EvollaConfig model) - [Exaone4_5_Config](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Config) configuration class: [Exaone4_5_ForConditionalGeneration](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_ForConditionalGeneration) (Exaone4_5_Config model) - [FastVlmConfig](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmConfig) configuration class: [FastVlmForConditionalGeneration](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmForConditionalGeneration) (FastVlmConfig model) - [Florence2Config](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Config) configuration class: [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (Florence2Config model) - [FunAsrNanoConfig](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoConfig) configuration class: [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (FunAsrNanoConfig model) - [FuyuConfig](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuConfig) configuration class: [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (FuyuConfig model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (Gemma3Config model) - [Gemma3nConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nConfig) configuration class: [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (Gemma3nConfig model) - [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) configuration class: [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (Gemma4Config model) - [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) configuration class: [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (Gemma4UnifiedConfig model) - [GitConfig](/docs/transformers/main/id/model_doc/git#transformers.GitConfig) configuration class: [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (GitConfig model) - [Glm46VConfig](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VConfig) configuration class: [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (Glm46VConfig model) - [Glm4vConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vConfig) configuration class: [Glm4vForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vForConditionalGeneration) (Glm4vConfig model) - [Glm4vMoeConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeConfig) configuration class: [Glm4vMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeForConditionalGeneration) (Glm4vMoeConfig model) - [Glm5NextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextConfig) configuration class: [Glm5NextForConditionalGeneration](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextForConditionalGeneration) (Glm5NextConfig model) - [GlmAsrConfig](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrConfig) configuration class: [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (GlmAsrConfig model) - [GlmOcrConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrConfig) configuration class: [GlmOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrForConditionalGeneration) (GlmOcrConfig model) - [GlmgaConfig](/docs/transformers/main/id/model_doc/glmga#transformers.GlmgaConfig) configuration class: [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (GlmgaConfig model) - [GotOcr2Config](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Config) configuration class: [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (GotOcr2Config model) - [Granite4VisionConfig](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionConfig) configuration class: [Granite4VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionForConditionalGeneration) (Granite4VisionConfig model) - [GraniteSpeechConfig](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechConfig) configuration class: [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (GraniteSpeechConfig model) - [GraniteSpeechPlusConfig](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusConfig) configuration class: [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (GraniteSpeechPlusConfig model) - [HunYuanVLConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLConfig) configuration class: [HunYuanVLForConditionalGeneration](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLForConditionalGeneration) (HunYuanVLConfig model) - [HyperCLOVAXVisionV2Config](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Config) configuration class: [HyperCLOVAXVisionV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2ForConditionalGeneration) (HyperCLOVAXVisionV2Config model) - [Idefics2Config](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Config) configuration class: [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (Idefics2Config model) - [Idefics3Config](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Config) configuration class: [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (Idefics3Config model) - [IdeficsConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsConfig) configuration class: [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (IdeficsConfig model) - [InklingConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingConfig) configuration class: [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (InklingConfig model) - [InstructBlipConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipConfig) configuration class: [InstructBlipForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipForConditionalGeneration) (InstructBlipConfig model) - [InstructBlipVideoConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoConfig) configuration class: [InstructBlipVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoForConditionalGeneration) (InstructBlipVideoConfig model) - [InternVLConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLConfig) configuration class: [InternVLForConditionalGeneration](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLForConditionalGeneration) (InternVLConfig model) - [JanusConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusConfig) configuration class: [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (JanusConfig model) - [Kimi_K25Config](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Config) configuration class: [Kimi_K25ForConditionalGeneration](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25ForConditionalGeneration) (Kimi_K25Config model) - [Kosmos2Config](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Config) configuration class: [Kosmos2ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2ForConditionalGeneration) (Kosmos2Config model) - [Kosmos2_5Config](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Config) configuration class: [Kosmos2_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5ForConditionalGeneration) (Kosmos2_5Config model) - [KyutaiSpeechToTextConfig](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextConfig) configuration class: [KyutaiSpeechToTextForConditionalGeneration](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextForConditionalGeneration) (KyutaiSpeechToTextConfig model) - [Lfm2VlConfig](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlConfig) configuration class: [Lfm2VlForConditionalGeneration](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlForConditionalGeneration) (Lfm2VlConfig model) - [LightOnOcrConfig](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrConfig) configuration class: [LightOnOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrForConditionalGeneration) (LightOnOcrConfig model) - [Llama4Config](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Config) configuration class: [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (Llama4Config model) - [LlavaConfig](/docs/transformers/main/id/model_doc/llava#transformers.LlavaConfig) configuration class: [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (LlavaConfig model) - [LlavaNextConfig](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextConfig) configuration class: [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (LlavaNextConfig model) - [LlavaNextVideoConfig](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoConfig) configuration class: [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (LlavaNextVideoConfig model) - [LlavaOnevisionConfig](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionConfig) configuration class: [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (LlavaOnevisionConfig model) - [MiniCPMOcrConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrConfig) configuration class: [MiniCPMOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrForConditionalGeneration) (MiniCPMOcrConfig model) - [MiniCPMV4_6Config](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Config) configuration class: [MiniCPMV4_6ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6ForConditionalGeneration) (MiniCPMV4_6Config model) - [MiniCPMV4_7Config](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Config) configuration class: [MiniCPMV4_7ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7ForConditionalGeneration) (MiniCPMV4_7Config model) - [MiniMaxM3VLConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLConfig) configuration class: [MiniMaxM3SparseForConditionalGeneration](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3SparseForConditionalGeneration) (MiniMaxM3VLConfig model) - [Mistral3Config](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Config) configuration class: [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (Mistral3Config model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (Mistral4Config model) - [MistralLarge4Config](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Config) configuration class: [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (MistralLarge4Config model) - [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) configuration class: [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (MllamaConfig model) - [Molmo2Config](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Config) configuration class: [Molmo2ForConditionalGeneration](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2ForConditionalGeneration) (Molmo2Config model) - [MossTranscribeDiarizeConfig](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeConfig) configuration class: [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (MossTranscribeDiarizeConfig model) - [MuseGlimmerConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerConfig) configuration class: [MuseGlimmerForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerForConditionalGeneration) (MuseGlimmerConfig model) - [MuseSparkConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkConfig) configuration class: [MuseSparkForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkForConditionalGeneration) (MuseSparkConfig model) - [Nemotron3_5TranscribeConfig](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeConfig) configuration class: [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (Nemotron3_5TranscribeConfig model) - [NemotronH_Omni_Reasoning_V3_Config](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3_Config) configuration class: [NemotronH_Omni_Reasoning_V3](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3) (NemotronH_Omni_Reasoning_V3_Config model) - [Ovis2Config](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Config) configuration class: [Ovis2ForConditionalGeneration](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2ForConditionalGeneration) (Ovis2Config model) - [PI0Config](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Config) configuration class: [PI0ForConditionalGeneration](/docs/transformers/main/id/model_doc/pi0#transformers.PI0ForConditionalGeneration) (PI0Config model) - [PPChart2TableConfig](/docs/transformers/main/id/model_doc/pp_chart2table#transformers.PPChart2TableConfig) configuration class: [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (PPChart2TableConfig model) - [PPFormulaNetConfig](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetConfig) configuration class: [PPFormulaNetForConditionalGeneration](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetForConditionalGeneration) (PPFormulaNetConfig model) - [PaddleOCRVLConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLConfig) configuration class: [PaddleOCRVLForConditionalGeneration](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLForConditionalGeneration) (PaddleOCRVLConfig model) - [PaliGemmaConfig](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaConfig) configuration class: [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (PaliGemmaConfig model) - [PerceptionLMConfig](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMConfig) configuration class: [PerceptionLMForConditionalGeneration](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMForConditionalGeneration) (PerceptionLMConfig model) - [Phi4MultimodalConfig](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalConfig) configuration class: [Phi4MultimodalForCausalLM](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalForCausalLM) (Phi4MultimodalConfig model) - [Pix2StructConfig](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructConfig) configuration class: [Pix2StructForConditionalGeneration](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructForConditionalGeneration) (Pix2StructConfig model) - [QianfanOCRConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRConfig) configuration class: [QianfanOCRForConditionalGeneration](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRForConditionalGeneration) (QianfanOCRConfig model) - [Qwen2AudioConfig](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioConfig) configuration class: [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (Qwen2AudioConfig model) - [Qwen2VLConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLConfig) configuration class: [Qwen2VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLForConditionalGeneration) (Qwen2VLConfig model) - [Qwen2_5OmniConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniConfig) configuration class: [Qwen2_5OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniForConditionalGeneration) (Qwen2_5OmniConfig model) - [Qwen2_5OmniThinkerConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerConfig) configuration class: [Qwen2_5OmniThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerForConditionalGeneration) (Qwen2_5OmniThinkerConfig model) - [Qwen2_5_VLConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLConfig) configuration class: [Qwen2_5_VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLForConditionalGeneration) (Qwen2_5_VLConfig model) - [Qwen3ASRConfig](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRConfig) configuration class: [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (Qwen3ASRConfig model) - [Qwen3OmniMoeConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeConfig) configuration class: [Qwen3OmniMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeForConditionalGeneration) (Qwen3OmniMoeConfig model) - [Qwen3OmniMoeThinkerConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerConfig) configuration class: [Qwen3OmniMoeThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerForConditionalGeneration) (Qwen3OmniMoeThinkerConfig model) - [Qwen3VLConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLConfig) configuration class: [Qwen3VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLForConditionalGeneration) (Qwen3VLConfig model) - [Qwen3VLMoeConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeConfig) configuration class: [Qwen3VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeForConditionalGeneration) (Qwen3VLMoeConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForConditionalGeneration) (Qwen3_5Config model) - [Qwen3_5MoeConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeConfig) configuration class: [Qwen3_5MoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForConditionalGeneration) (Qwen3_5MoeConfig model) - [Qwen4ExpConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpConfig) configuration class: [Qwen4ExpForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForConditionalGeneration) (Qwen4ExpConfig model) - [ShieldGemma2Config](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (ShieldGemma2Config model) - [SmolVLMConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMConfig) configuration class: [SmolVLMForConditionalGeneration](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration) (SmolVLMConfig model) - [Step3p7Config](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Config) configuration class: [Step3p7ForConditionalGeneration](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7ForConditionalGeneration) (Step3p7Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (T5Gemma2Config model) - [UdopConfig](/docs/transformers/main/id/model_doc/udop#transformers.UdopConfig) configuration class: [UdopForConditionalGeneration](/docs/transformers/main/id/model_doc/udop#transformers.UdopForConditionalGeneration) (UdopConfig model) - [UnlimitedOcrConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrConfig) configuration class: [UnlimitedOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrForConditionalGeneration) (UnlimitedOcrConfig model) - [VibeVoiceAsrConfig](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrConfig) configuration class: [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (VibeVoiceAsrConfig model) - [VideoLlama3Config](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Config) configuration class: [VideoLlama3ForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3ForConditionalGeneration) (VideoLlama3Config model) - [VideoLlavaConfig](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaConfig) configuration class: [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (VideoLlavaConfig model) - [VipLlavaConfig](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaConfig) configuration class: [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (VipLlavaConfig model) - [VisionEncoderDecoderConfig](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderConfig) configuration class: [VisionEncoderDecoderModel](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderModel) (VisionEncoderDecoderConfig model) - [VoxtralConfig](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralConfig) configuration class: [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (VoxtralConfig model) - [VoxtralRealtimeConfig](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeConfig) configuration class: [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (VoxtralRealtimeConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala generasi multimodal) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMultimodalLM

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForMultimodalLM.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForMultimodalLM.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala generasi multimodal) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **aria** -- [AriaForConditionalGeneration](/docs/transformers/main/id/model_doc/aria#transformers.AriaForConditionalGeneration) (model AriaConfig)
- **aya_vision** -- [AyaVisionForConditionalGeneration](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionForConditionalGeneration) (model AyaVisionConfig)
- **blip** -- [BlipForConditionalGeneration](/docs/transformers/main/id/model_doc/blip#transformers.BlipForConditionalGeneration) (model BlipConfig)
- **blip-2** -- [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (model Blip2Config)
- **chameleon** -- [ChameleonForConditionalGeneration](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonForConditionalGeneration) (model ChameleonConfig)
- **cohere2_vision** -- [Cohere2VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionForConditionalGeneration) (model Cohere2VisionConfig)
- **cohere_compass** -- [CohereCompassForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForConditionalGeneration) (model CohereCompassConfig)
- **cosmos3_edge** -- [Cosmos3EdgeForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeForConditionalGeneration) (model Cosmos3EdgeConfig)
- **cosmos3_omni** -- [Cosmos3OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniForConditionalGeneration) (model Cosmos3OmniConfig)
- **deepseek_ocr2** -- [DeepseekOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2ForConditionalGeneration) (model DeepseekOcr2Config)
- **deepseek_vl** -- [DeepseekVLForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLForConditionalGeneration) (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- [DeepseekVLHybridForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridForConditionalGeneration) (model DeepseekVLHybridConfig)
- **diffusion_gemma** -- [DiffusionGemmaForBlockDiffusion](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaForBlockDiffusion) (model DiffusionGemmaConfig)
- **emu3** -- [Emu3ForConditionalGeneration](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForConditionalGeneration) (model Emu3Config)
- **ernie4_5_vl_moe** -- [Ernie4_5_VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeForConditionalGeneration) (model Ernie4_5_VLMoeConfig)
- **evolla** -- [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (model EvollaConfig)
- **exaone4_5** -- [Exaone4_5_ForConditionalGeneration](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_ForConditionalGeneration) (model Exaone4_5_Config)
- **fast_vlm** -- [FastVlmForConditionalGeneration](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmForConditionalGeneration) (model FastVlmConfig)
- **florence2** -- [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (model Florence2Config)
- **fun_asr_nano** -- [FunAsrNanoForConditionalGeneration](/docs/transformers/main/id/model_doc/fun_asr_nano#transformers.FunAsrNanoForConditionalGeneration) (model FunAsrNanoConfig)
- **fuyu** -- [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (model FuyuConfig)
- **gemma3** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model Gemma3Config)
- **gemma3n** -- [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (model Gemma3nConfig)
- **gemma4** -- [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (model Gemma4Config)
- **gemma4_unified** -- [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (model Gemma4UnifiedConfig)
- **git** -- [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (model GitConfig)
- **glm46v** -- [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (model Glm46VConfig)
- **glm4v** -- [Glm4vForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vForConditionalGeneration) (model Glm4vConfig)
- **glm4v_moe** -- [Glm4vMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeForConditionalGeneration) (model Glm4vMoeConfig)
- **glm5_next** -- [Glm5NextForConditionalGeneration](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextForConditionalGeneration) (model Glm5NextConfig)
- **glm_ocr** -- [GlmOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrForConditionalGeneration) (model GlmOcrConfig)
- **glmasr** -- [GlmAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/glmasr#transformers.GlmAsrForConditionalGeneration) (model GlmAsrConfig)
- **glmga** -- [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (model GlmgaConfig)
- **got_ocr2** -- [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (model GotOcr2Config)
- **granite4_vision** -- [Granite4VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionForConditionalGeneration) (model Granite4VisionConfig)
- **granite_speech** -- [GraniteSpeechForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech#transformers.GraniteSpeechForConditionalGeneration) (model GraniteSpeechConfig)
- **granite_speech_plus** -- [GraniteSpeechPlusForConditionalGeneration](/docs/transformers/main/id/model_doc/granite_speech_plus#transformers.GraniteSpeechPlusForConditionalGeneration) (model GraniteSpeechPlusConfig)
- **hunyuan_vl** -- [HunYuanVLForConditionalGeneration](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLForConditionalGeneration) (model HunYuanVLConfig)
- **hyperclovax_vision_v2** -- [HyperCLOVAXVisionV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2ForConditionalGeneration) (model HyperCLOVAXVisionV2Config)
- **idefics** -- [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (model IdeficsConfig)
- **idefics2** -- [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (model Idefics2Config)
- **idefics3** -- [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (model Idefics3Config)
- **inkling_mm_model** -- [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (model InklingConfig)
- **instructblip** -- [InstructBlipForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipForConditionalGeneration) (model InstructBlipConfig)
- **instructblipvideo** -- [InstructBlipVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoForConditionalGeneration) (model InstructBlipVideoConfig)
- **internvl** -- [InternVLForConditionalGeneration](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLForConditionalGeneration) (model InternVLConfig)
- **janus** -- [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (model JanusConfig)
- **kimi_k25** -- [Kimi_K25ForConditionalGeneration](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25ForConditionalGeneration) (model Kimi_K25Config)
- **kosmos-2** -- [Kosmos2ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2ForConditionalGeneration) (model Kosmos2Config)
- **kosmos-2.5** -- [Kosmos2_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5ForConditionalGeneration) (model Kosmos2_5Config)
- **kyutai_speech_to_text** -- [KyutaiSpeechToTextForConditionalGeneration](/docs/transformers/main/id/model_doc/kyutai_speech_to_text#transformers.KyutaiSpeechToTextForConditionalGeneration) (model KyutaiSpeechToTextConfig)
- **lfm2_vl** -- [Lfm2VlForConditionalGeneration](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlForConditionalGeneration) (model Lfm2VlConfig)
- **lighton_ocr** -- [LightOnOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrForConditionalGeneration) (model LightOnOcrConfig)
- **llama4** -- [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (model Llama4Config)
- **llava** -- [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (model LlavaConfig)
- **llava_next** -- [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (model LlavaOnevisionConfig)
- **minicpm_ocr** -- [MiniCPMOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrForConditionalGeneration) (model MiniCPMOcrConfig)
- **minicpmv4_6** -- [MiniCPMV4_6ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6ForConditionalGeneration) (model MiniCPMV4_6Config)
- **minicpmv4_7** -- [MiniCPMV4_7ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7ForConditionalGeneration) (model MiniCPMV4_7Config)
- **minimax_m3_vl** -- [MiniMaxM3SparseForConditionalGeneration](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3SparseForConditionalGeneration) (model MiniMaxM3VLConfig)
- **mistral3** -- [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (model Mistral3Config)
- **mistral4** -- [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (model Mistral4Config)
- **mistral_large4** -- [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (model MistralLarge4Config)
- **mllama** -- [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (model MllamaConfig)
- **molmo2** -- [Molmo2ForConditionalGeneration](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2ForConditionalGeneration) (model Molmo2Config)
- **moss_transcribe_diarize** -- [MossTranscribeDiarizeForConditionalGeneration](/docs/transformers/main/id/model_doc/moss_transcribe_diarize#transformers.MossTranscribeDiarizeForConditionalGeneration) (model MossTranscribeDiarizeConfig)
- **muse_glimmer** -- [MuseGlimmerForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerForConditionalGeneration) (model MuseGlimmerConfig)
- **muse_spark** -- [MuseSparkForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkForConditionalGeneration) (model MuseSparkConfig)
- **nemo_speechlm** -- [Nemotron3_5TranscribeForConditionalGeneration](/docs/transformers/main/id/model_doc/nemotron3_5_transcribe#transformers.Nemotron3_5TranscribeForConditionalGeneration) (model Nemotron3_5TranscribeConfig)
- **nemotron_h_omni** -- [NemotronH_Omni_Reasoning_V3](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3) (model NemotronH_Omni_Reasoning_V3_Config)
- **ovis2** -- [Ovis2ForConditionalGeneration](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2ForConditionalGeneration) (model Ovis2Config)
- **paddleocr_vl** -- [PaddleOCRVLForConditionalGeneration](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLForConditionalGeneration) (model PaddleOCRVLConfig)
- **paligemma** -- [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (model PaliGemmaConfig)
- **perception_lm** -- [PerceptionLMForConditionalGeneration](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMForConditionalGeneration) (model PerceptionLMConfig)
- **phi4_multimodal** -- [Phi4MultimodalForCausalLM](/docs/transformers/main/id/model_doc/phi4_multimodal#transformers.Phi4MultimodalForCausalLM) (model Phi4MultimodalConfig)
- **pi0** -- [PI0ForConditionalGeneration](/docs/transformers/main/id/model_doc/pi0#transformers.PI0ForConditionalGeneration) (model PI0Config)
- **pix2struct** -- [Pix2StructForConditionalGeneration](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructForConditionalGeneration) (model Pix2StructConfig)
- **pp_chart2table** -- [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (model PPChart2TableConfig)
- **pp_formulanet** -- [PPFormulaNetForConditionalGeneration](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetForConditionalGeneration) (model PPFormulaNetConfig)
- **qianfan_ocr** -- [QianfanOCRForConditionalGeneration](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRForConditionalGeneration) (model QianfanOCRConfig)
- **qwen2_5_omni** -- [Qwen2_5OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniForConditionalGeneration) (model Qwen2_5OmniConfig)
- **qwen2_5_omni_thinker** -- [Qwen2_5OmniThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerForConditionalGeneration) (model Qwen2_5OmniThinkerConfig)
- **qwen2_5_vl** -- [Qwen2_5_VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLForConditionalGeneration) (model Qwen2_5_VLConfig)
- **qwen2_audio** -- [Qwen2AudioForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_audio#transformers.Qwen2AudioForConditionalGeneration) (model Qwen2AudioConfig)
- **qwen2_vl** -- [Qwen2VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLForConditionalGeneration) (model Qwen2VLConfig)
- **qwen3_5** -- [Qwen3_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForConditionalGeneration) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5MoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForConditionalGeneration) (model Qwen3_5MoeConfig)
- **qwen3_asr** -- [Qwen3ASRForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_asr#transformers.Qwen3ASRForConditionalGeneration) (model Qwen3ASRConfig)
- **qwen3_omni_moe** -- [Qwen3OmniMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeForConditionalGeneration) (model Qwen3OmniMoeConfig)
- **qwen3_omni_moe_thinker** -- [Qwen3OmniMoeThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerForConditionalGeneration) (model Qwen3OmniMoeThinkerConfig)
- **qwen3_vl** -- [Qwen3VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLForConditionalGeneration) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen3VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeForConditionalGeneration) (model Qwen3VLMoeConfig)
- **qwen4_exp** -- [Qwen4ExpForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForConditionalGeneration) (model Qwen4ExpConfig)
- **shieldgemma2** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model ShieldGemma2Config)
- **smolvlm** -- [SmolVLMForConditionalGeneration](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration) (model SmolVLMConfig)
- **step3p7** -- [Step3p7ForConditionalGeneration](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7ForConditionalGeneration) (model Step3p7Config)
- **t5gemma2** -- [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (model T5Gemma2Config)
- **udop** -- [UdopForConditionalGeneration](/docs/transformers/main/id/model_doc/udop#transformers.UdopForConditionalGeneration) (model UdopConfig)
- **unlimited_ocr** -- [UnlimitedOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrForConditionalGeneration) (model UnlimitedOcrConfig)
- **vibevoice_asr** -- [VibeVoiceAsrForConditionalGeneration](/docs/transformers/main/id/model_doc/vibevoice_asr#transformers.VibeVoiceAsrForConditionalGeneration) (model VibeVoiceAsrConfig)
- **video_llama_3** -- [VideoLlama3ForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3ForConditionalGeneration) (model VideoLlama3Config)
- **video_llava** -- [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (model VideoLlavaConfig)
- **vipllava** -- [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (model VipLlavaConfig)
- **vision-encoder-decoder** -- [VisionEncoderDecoderModel](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderModel) (model VisionEncoderDecoderConfig)
- **voxtral** -- [VoxtralForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral#transformers.VoxtralForConditionalGeneration) (model VoxtralConfig)
- **voxtral_realtime** -- [VoxtralRealtimeForConditionalGeneration](/docs/transformers/main/id/model_doc/voxtral_realtime#transformers.VoxtralRealtimeForConditionalGeneration) (model VoxtralRealtimeConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForMultimodalLM

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForMultimodalLM.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForMultimodalLM.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForTableQuestionAnswering[[transformers.AutoModelForTableQuestionAnswering]]

#### transformers.AutoModelForTableQuestionAnswering[[transformers.AutoModelForTableQuestionAnswering]]

```python
transformers.AutoModelForTableQuestionAnswering(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2362)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala penjawab pertanyaan tabel) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTableQuestionAnswering.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi: - Kelas konfigurasi [TapasConfig](/docs/transformers/main/id/model_doc/tapas#transformers.TapasConfig): [TapasForQuestionAnswering](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForQuestionAnswering) (model TapasConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala tanya jawab tabel) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTableQuestionAnswering

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google/tapas-base-finetuned-wtq")
>>> model = AutoModelForTableQuestionAnswering.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTableQuestionAnswering.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala table question answering) dari model pretrained.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **tapas** -- [TapasForQuestionAnswering](/docs/transformers/main/id/model_doc/tapas#transformers.TapasForQuestionAnswering) (model TapasConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTableQuestionAnswering

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTableQuestionAnswering.from_pretrained("google/tapas-base-finetuned-wtq")

>>> # Update configuration during loading
>>> model = AutoModelForTableQuestionAnswering.from_pretrained("google/tapas-base-finetuned-wtq", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForDocumentQuestionAnswering[[transformers.AutoModelForDocumentQuestionAnswering]]

#### transformers.AutoModelForDocumentQuestionAnswering[[transformers.AutoModelForDocumentQuestionAnswering]]

```python
transformers.AutoModelForDocumentQuestionAnswering(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2384)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala untuk menjawab pertanyaan dokumen) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForDocumentQuestionAnswering.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [LayoutLMConfig](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMConfig): [LayoutLMForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForQuestionAnswering) (model LayoutLMConfig) - Kelas konfigurasi [LayoutLMv2Config](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2Config): [LayoutLMv2ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForQuestionAnswering) (model LayoutLMv2Config) - Kelas konfigurasi [LayoutLMv3Config](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3Config): [LayoutLMv3ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForQuestionAnswering) (model LayoutLMv3Config)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala untuk menjawab pertanyaan dokumen) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForDocumentQuestionAnswering

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("impira/layoutlm-document-qa", revision="52e01b3")
>>> model = AutoModelForDocumentQuestionAnswering.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForDocumentQuestionAnswering.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala untuk menjawab pertanyaan dokumen) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **layoutlm** -- [LayoutLMForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlm#transformers.LayoutLMForQuestionAnswering) (model LayoutLMConfig)
- **layoutlmv2** -- [LayoutLMv2ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv2#transformers.LayoutLMv2ForQuestionAnswering) (model LayoutLMv2Config)
- **layoutlmv3** -- [LayoutLMv3ForQuestionAnswering](/docs/transformers/main/id/model_doc/layoutlmv3#transformers.LayoutLMv3ForQuestionAnswering) (model LayoutLMv3Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForDocumentQuestionAnswering

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForDocumentQuestionAnswering.from_pretrained("impira/layoutlm-document-qa", revision="52e01b3")

>>> # Update configuration during loading
>>> model = AutoModelForDocumentQuestionAnswering.from_pretrained("impira/layoutlm-document-qa", revision="52e01b3", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForVisualQuestionAnswering[[transformers.AutoModelForVisualQuestionAnswering]]

#### transformers.AutoModelForVisualQuestionAnswering[[transformers.AutoModelForVisualQuestionAnswering]]

```python
transformers.AutoModelForVisualQuestionAnswering(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2373)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala visual question answering) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForVisualQuestionAnswering.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config): [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (model Blip2Config) - Kelas konfigurasi [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig): [BlipForQuestionAnswering](/docs/transformers/main/id/model_doc/blip#transformers.BlipForQuestionAnswering) (model BlipConfig) - Kelas konfigurasi [ViltConfig](/docs/transformers/main/id/model_doc/vilt#transformers.ViltConfig): [ViltForQuestionAnswering](/docs/transformers/main/id/model_doc/vilt#transformers.ViltForQuestionAnswering) (model ViltConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala visual question answering) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForVisualQuestionAnswering

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("dandelin/vilt-b32-finetuned-vqa")
>>> model = AutoModelForVisualQuestionAnswering.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForVisualQuestionAnswering.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala visual question answering) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **blip** -- [BlipForQuestionAnswering](/docs/transformers/main/id/model_doc/blip#transformers.BlipForQuestionAnswering) (model BlipConfig)
- **blip-2** -- [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (model Blip2Config)
- **vilt** -- [ViltForQuestionAnswering](/docs/transformers/main/id/model_doc/vilt#transformers.ViltForQuestionAnswering) (model ViltConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForVisualQuestionAnswering

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForVisualQuestionAnswering.from_pretrained("dandelin/vilt-b32-finetuned-vqa")

>>> # Update configuration during loading
>>> model = AutoModelForVisualQuestionAnswering.from_pretrained("dandelin/vilt-b32-finetuned-vqa", output_attentions=True)
>>> model.config.output_attentions
True
```

### AutoModelForImageTextToText[[transformers.AutoModelForImageTextToText]]

#### transformers.AutoModelForImageTextToText[[transformers.AutoModelForImageTextToText]]

```python
transformers.AutoModelForImageTextToText(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2521)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala pemodelan gambar-teks-ke-teks) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForImageTextToText.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : The model class to instantiate is selected based on the configuration class:  - [AriaConfig](/docs/transformers/main/id/model_doc/aria#transformers.AriaConfig) configuration class: [AriaForConditionalGeneration](/docs/transformers/main/id/model_doc/aria#transformers.AriaForConditionalGeneration) (AriaConfig model) - [AyaVisionConfig](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionConfig) configuration class: [AyaVisionForConditionalGeneration](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionForConditionalGeneration) (AyaVisionConfig model) - [Blip2Config](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2Config) configuration class: [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (Blip2Config model) - [BlipConfig](/docs/transformers/main/id/model_doc/blip#transformers.BlipConfig) configuration class: [BlipForConditionalGeneration](/docs/transformers/main/id/model_doc/blip#transformers.BlipForConditionalGeneration) (BlipConfig model) - [ChameleonConfig](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonConfig) configuration class: [ChameleonForConditionalGeneration](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonForConditionalGeneration) (ChameleonConfig model) - [Cohere2VisionConfig](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionConfig) configuration class: [Cohere2VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionForConditionalGeneration) (Cohere2VisionConfig model) - [CohereCompassConfig](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassConfig) configuration class: [CohereCompassForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForConditionalGeneration) (CohereCompassConfig model) - [Cosmos3EdgeConfig](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeConfig) configuration class: [Cosmos3EdgeForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeForConditionalGeneration) (Cosmos3EdgeConfig model) - [Cosmos3OmniConfig](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniConfig) configuration class: [Cosmos3OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniForConditionalGeneration) (Cosmos3OmniConfig model) - [DeepseekOcr2Config](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2Config) configuration class: [DeepseekOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2ForConditionalGeneration) (DeepseekOcr2Config model) - [DeepseekVLConfig](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLConfig) configuration class: [DeepseekVLForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLForConditionalGeneration) (DeepseekVLConfig model) - [DeepseekVLHybridConfig](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridConfig) configuration class: [DeepseekVLHybridForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridForConditionalGeneration) (DeepseekVLHybridConfig model) - [DiffusionGemmaConfig](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaConfig) configuration class: [DiffusionGemmaForBlockDiffusion](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaForBlockDiffusion) (DiffusionGemmaConfig model) - [Emu3Config](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3Config) configuration class: [Emu3ForConditionalGeneration](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForConditionalGeneration) (Emu3Config model) - [Ernie4_5_VLMoeConfig](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeConfig) configuration class: [Ernie4_5_VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeForConditionalGeneration) (Ernie4_5_VLMoeConfig model) - [EvollaConfig](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaConfig) configuration class: [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (EvollaConfig model) - [Exaone4_5_Config](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_Config) configuration class: [Exaone4_5_ForConditionalGeneration](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_ForConditionalGeneration) (Exaone4_5_Config model) - [FastVlmConfig](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmConfig) configuration class: [FastVlmForConditionalGeneration](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmForConditionalGeneration) (FastVlmConfig model) - [Florence2Config](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2Config) configuration class: [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (Florence2Config model) - [FuyuConfig](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuConfig) configuration class: [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (FuyuConfig model) - [Gemma3Config](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (Gemma3Config model) - [Gemma3nConfig](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nConfig) configuration class: [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (Gemma3nConfig model) - [Gemma4Config](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4Config) configuration class: [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (Gemma4Config model) - [Gemma4UnifiedConfig](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedConfig) configuration class: [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (Gemma4UnifiedConfig model) - [GitConfig](/docs/transformers/main/id/model_doc/git#transformers.GitConfig) configuration class: [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (GitConfig model) - [Glm46VConfig](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VConfig) configuration class: [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (Glm46VConfig model) - [Glm4vConfig](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vConfig) configuration class: [Glm4vForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vForConditionalGeneration) (Glm4vConfig model) - [Glm4vMoeConfig](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeConfig) configuration class: [Glm4vMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeForConditionalGeneration) (Glm4vMoeConfig model) - [Glm5NextConfig](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextConfig) configuration class: [Glm5NextForConditionalGeneration](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextForConditionalGeneration) (Glm5NextConfig model) - [GlmOcrConfig](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrConfig) configuration class: [GlmOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrForConditionalGeneration) (GlmOcrConfig model) - [GlmgaConfig](/docs/transformers/main/id/model_doc/glmga#transformers.GlmgaConfig) configuration class: [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (GlmgaConfig model) - [GotOcr2Config](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2Config) configuration class: [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (GotOcr2Config model) - [Granite4VisionConfig](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionConfig) configuration class: [Granite4VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionForConditionalGeneration) (Granite4VisionConfig model) - [HunYuanVLConfig](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLConfig) configuration class: [HunYuanVLForConditionalGeneration](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLForConditionalGeneration) (HunYuanVLConfig model) - [HyperCLOVAXVisionV2Config](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2Config) configuration class: [HyperCLOVAXVisionV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2ForConditionalGeneration) (HyperCLOVAXVisionV2Config model) - [Idefics2Config](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2Config) configuration class: [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (Idefics2Config model) - [Idefics3Config](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3Config) configuration class: [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (Idefics3Config model) - [IdeficsConfig](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsConfig) configuration class: [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (IdeficsConfig model) - [InklingConfig](/docs/transformers/main/id/model_doc/inkling#transformers.InklingConfig) configuration class: [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (InklingConfig model) - [InstructBlipConfig](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipConfig) configuration class: [InstructBlipForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipForConditionalGeneration) (InstructBlipConfig model) - [InstructBlipVideoConfig](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoConfig) configuration class: [InstructBlipVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoForConditionalGeneration) (InstructBlipVideoConfig model) - [InternVLConfig](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLConfig) configuration class: [InternVLForConditionalGeneration](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLForConditionalGeneration) (InternVLConfig model) - [JanusConfig](/docs/transformers/main/id/model_doc/janus#transformers.JanusConfig) configuration class: [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (JanusConfig model) - [Kimi_K25Config](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25Config) configuration class: [Kimi_K25ForConditionalGeneration](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25ForConditionalGeneration) (Kimi_K25Config model) - [Kosmos2Config](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2Config) configuration class: [Kosmos2ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2ForConditionalGeneration) (Kosmos2Config model) - [Kosmos2_5Config](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5Config) configuration class: [Kosmos2_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5ForConditionalGeneration) (Kosmos2_5Config model) - [Lfm2VlConfig](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlConfig) configuration class: [Lfm2VlForConditionalGeneration](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlForConditionalGeneration) (Lfm2VlConfig model) - [LightOnOcrConfig](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrConfig) configuration class: [LightOnOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrForConditionalGeneration) (LightOnOcrConfig model) - [Llama4Config](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4Config) configuration class: [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (Llama4Config model) - [LlavaConfig](/docs/transformers/main/id/model_doc/llava#transformers.LlavaConfig) configuration class: [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (LlavaConfig model) - [LlavaNextConfig](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextConfig) configuration class: [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (LlavaNextConfig model) - [LlavaNextVideoConfig](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoConfig) configuration class: [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (LlavaNextVideoConfig model) - [LlavaOnevisionConfig](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionConfig) configuration class: [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (LlavaOnevisionConfig model) - [MiniCPMOcrConfig](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrConfig) configuration class: [MiniCPMOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrForConditionalGeneration) (MiniCPMOcrConfig model) - [MiniCPMV4_6Config](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6Config) configuration class: [MiniCPMV4_6ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6ForConditionalGeneration) (MiniCPMV4_6Config model) - [MiniCPMV4_7Config](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7Config) configuration class: [MiniCPMV4_7ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7ForConditionalGeneration) (MiniCPMV4_7Config model) - [MiniMaxM3VLConfig](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3VLConfig) configuration class: [MiniMaxM3SparseForConditionalGeneration](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3SparseForConditionalGeneration) (MiniMaxM3VLConfig model) - [Mistral3Config](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3Config) configuration class: [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (Mistral3Config model) - [Mistral4Config](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4Config) configuration class: [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (Mistral4Config model) - [MistralLarge4Config](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4Config) configuration class: [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (MistralLarge4Config model) - [MllamaConfig](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaConfig) configuration class: [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (MllamaConfig model) - [Molmo2Config](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2Config) configuration class: [Molmo2ForConditionalGeneration](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2ForConditionalGeneration) (Molmo2Config model) - [MuseGlimmerConfig](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerConfig) configuration class: [MuseGlimmerForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerForConditionalGeneration) (MuseGlimmerConfig model) - [MuseSparkConfig](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkConfig) configuration class: [MuseSparkForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkForConditionalGeneration) (MuseSparkConfig model) - [NemotronH_Omni_Reasoning_V3_Config](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3_Config) configuration class: [NemotronH_Omni_Reasoning_V3](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3) (NemotronH_Omni_Reasoning_V3_Config model) - [Ovis2Config](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2Config) configuration class: [Ovis2ForConditionalGeneration](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2ForConditionalGeneration) (Ovis2Config model) - [PI0Config](/docs/transformers/main/id/model_doc/pi0#transformers.PI0Config) configuration class: [PI0ForConditionalGeneration](/docs/transformers/main/id/model_doc/pi0#transformers.PI0ForConditionalGeneration) (PI0Config model) - [PPChart2TableConfig](/docs/transformers/main/id/model_doc/pp_chart2table#transformers.PPChart2TableConfig) configuration class: [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (PPChart2TableConfig model) - [PPFormulaNetConfig](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetConfig) configuration class: [PPFormulaNetForConditionalGeneration](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetForConditionalGeneration) (PPFormulaNetConfig model) - [PaddleOCRVLConfig](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLConfig) configuration class: [PaddleOCRVLForConditionalGeneration](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLForConditionalGeneration) (PaddleOCRVLConfig model) - [PaliGemmaConfig](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaConfig) configuration class: [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (PaliGemmaConfig model) - [PerceptionLMConfig](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMConfig) configuration class: [PerceptionLMForConditionalGeneration](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMForConditionalGeneration) (PerceptionLMConfig model) - [Pix2StructConfig](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructConfig) configuration class: [Pix2StructForConditionalGeneration](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructForConditionalGeneration) (Pix2StructConfig model) - [QianfanOCRConfig](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRConfig) configuration class: [QianfanOCRForConditionalGeneration](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRForConditionalGeneration) (QianfanOCRConfig model) - [Qwen2VLConfig](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLConfig) configuration class: [Qwen2VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLForConditionalGeneration) (Qwen2VLConfig model) - [Qwen2_5OmniThinkerConfig](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerConfig) configuration class: [Qwen2_5OmniThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerForConditionalGeneration) (Qwen2_5OmniThinkerConfig model) - [Qwen2_5_VLConfig](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLConfig) configuration class: [Qwen2_5_VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLForConditionalGeneration) (Qwen2_5_VLConfig model) - [Qwen3OmniMoeThinkerConfig](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerConfig) configuration class: [Qwen3OmniMoeThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerForConditionalGeneration) (Qwen3OmniMoeThinkerConfig model) - [Qwen3VLConfig](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLConfig) configuration class: [Qwen3VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLForConditionalGeneration) (Qwen3VLConfig model) - [Qwen3VLMoeConfig](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeConfig) configuration class: [Qwen3VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeForConditionalGeneration) (Qwen3VLMoeConfig model) - [Qwen3_5Config](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5Config) configuration class: [Qwen3_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForConditionalGeneration) (Qwen3_5Config model) - [Qwen3_5MoeConfig](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeConfig) configuration class: [Qwen3_5MoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForConditionalGeneration) (Qwen3_5MoeConfig model) - [Qwen4ExpConfig](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpConfig) configuration class: [Qwen4ExpForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForConditionalGeneration) (Qwen4ExpConfig model) - [ShieldGemma2Config](/docs/transformers/main/id/model_doc/shieldgemma2#transformers.ShieldGemma2Config) configuration class: [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (ShieldGemma2Config model) - [SmolVLMConfig](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMConfig) configuration class: [SmolVLMForConditionalGeneration](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration) (SmolVLMConfig model) - [Step3p7Config](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7Config) configuration class: [Step3p7ForConditionalGeneration](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7ForConditionalGeneration) (Step3p7Config model) - [T5Gemma2Config](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2Config) configuration class: [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (T5Gemma2Config model) - [UdopConfig](/docs/transformers/main/id/model_doc/udop#transformers.UdopConfig) configuration class: [UdopForConditionalGeneration](/docs/transformers/main/id/model_doc/udop#transformers.UdopForConditionalGeneration) (UdopConfig model) - [UnlimitedOcrConfig](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrConfig) configuration class: [UnlimitedOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrForConditionalGeneration) (UnlimitedOcrConfig model) - [VideoLlama3Config](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3Config) configuration class: [VideoLlama3ForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3ForConditionalGeneration) (VideoLlama3Config model) - [VideoLlavaConfig](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaConfig) configuration class: [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (VideoLlavaConfig model) - [VipLlavaConfig](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaConfig) configuration class: [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (VipLlavaConfig model) - [VisionEncoderDecoderConfig](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderConfig) configuration class: [VisionEncoderDecoderModel](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderModel) (VisionEncoderDecoderConfig model)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala pemodelan gambar-teks-ke-teks) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageTextToText

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForImageTextToText.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForImageTextToText.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala pemodelan gambar-teks-ke-teks) dari model terlatih sebelumnya.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **aria** -- [AriaForConditionalGeneration](/docs/transformers/main/id/model_doc/aria#transformers.AriaForConditionalGeneration) (model AriaConfig)
- **aya_vision** -- [AyaVisionForConditionalGeneration](/docs/transformers/main/id/model_doc/aya_vision#transformers.AyaVisionForConditionalGeneration) (model AyaVisionConfig)
- **blip** -- [BlipForConditionalGeneration](/docs/transformers/main/id/model_doc/blip#transformers.BlipForConditionalGeneration) (model BlipConfig)
- **blip-2** -- [Blip2ForConditionalGeneration](/docs/transformers/main/id/model_doc/blip-2#transformers.Blip2ForConditionalGeneration) (model Blip2Config)
- **chameleon** -- [ChameleonForConditionalGeneration](/docs/transformers/main/id/model_doc/chameleon#transformers.ChameleonForConditionalGeneration) (model ChameleonConfig)
- **cohere2_vision** -- [Cohere2VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere2_vision#transformers.Cohere2VisionForConditionalGeneration) (model Cohere2VisionConfig)
- **cohere_compass** -- [CohereCompassForConditionalGeneration](/docs/transformers/main/id/model_doc/cohere_compass#transformers.CohereCompassForConditionalGeneration) (model CohereCompassConfig)
- **cosmos3_edge** -- [Cosmos3EdgeForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_edge#transformers.Cosmos3EdgeForConditionalGeneration) (model Cosmos3EdgeConfig)
- **cosmos3_omni** -- [Cosmos3OmniForConditionalGeneration](/docs/transformers/main/id/model_doc/cosmos3_omni#transformers.Cosmos3OmniForConditionalGeneration) (model Cosmos3OmniConfig)
- **deepseek_ocr2** -- [DeepseekOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_ocr2#transformers.DeepseekOcr2ForConditionalGeneration) (model DeepseekOcr2Config)
- **deepseek_vl** -- [DeepseekVLForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl#transformers.DeepseekVLForConditionalGeneration) (model DeepseekVLConfig)
- **deepseek_vl_hybrid** -- [DeepseekVLHybridForConditionalGeneration](/docs/transformers/main/id/model_doc/deepseek_vl_hybrid#transformers.DeepseekVLHybridForConditionalGeneration) (model DeepseekVLHybridConfig)
- **diffusion_gemma** -- [DiffusionGemmaForBlockDiffusion](/docs/transformers/main/id/model_doc/diffusion_gemma#transformers.DiffusionGemmaForBlockDiffusion) (model DiffusionGemmaConfig)
- **emu3** -- [Emu3ForConditionalGeneration](/docs/transformers/main/id/model_doc/emu3#transformers.Emu3ForConditionalGeneration) (model Emu3Config)
- **ernie4_5_vl_moe** -- [Ernie4_5_VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/ernie4_5_vl_moe#transformers.Ernie4_5_VLMoeForConditionalGeneration) (model Ernie4_5_VLMoeConfig)
- **evolla** -- [EvollaForProteinText2Text](/docs/transformers/main/id/model_doc/evolla#transformers.EvollaForProteinText2Text) (model EvollaConfig)
- **exaone4_5** -- [Exaone4_5_ForConditionalGeneration](/docs/transformers/main/id/model_doc/exaone4_5#transformers.Exaone4_5_ForConditionalGeneration) (model Exaone4_5_Config)
- **fast_vlm** -- [FastVlmForConditionalGeneration](/docs/transformers/main/id/model_doc/fast_vlm#transformers.FastVlmForConditionalGeneration) (model FastVlmConfig)
- **florence2** -- [Florence2ForConditionalGeneration](/docs/transformers/main/id/model_doc/florence2#transformers.Florence2ForConditionalGeneration) (model Florence2Config)
- **fuyu** -- [FuyuForCausalLM](/docs/transformers/main/id/model_doc/fuyu#transformers.FuyuForCausalLM) (model FuyuConfig)
- **gemma3** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model Gemma3Config)
- **gemma3n** -- [Gemma3nForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3n#transformers.Gemma3nForConditionalGeneration) (model Gemma3nConfig)
- **gemma4** -- [Gemma4ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4#transformers.Gemma4ForConditionalGeneration) (model Gemma4Config)
- **gemma4_unified** -- [Gemma4UnifiedForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma4_unified#transformers.Gemma4UnifiedForConditionalGeneration) (model Gemma4UnifiedConfig)
- **git** -- [GitForCausalLM](/docs/transformers/main/id/model_doc/git#transformers.GitForCausalLM) (model GitConfig)
- **glm46v** -- [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (model Glm46VConfig)
- **glm4v** -- [Glm4vForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v#transformers.Glm4vForConditionalGeneration) (model Glm4vConfig)
- **glm4v_moe** -- [Glm4vMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/glm4v_moe#transformers.Glm4vMoeForConditionalGeneration) (model Glm4vMoeConfig)
- **glm5_next** -- [Glm5NextForConditionalGeneration](/docs/transformers/main/id/model_doc/glm5_next#transformers.Glm5NextForConditionalGeneration) (model Glm5NextConfig)
- **glm_ocr** -- [GlmOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/glm_ocr#transformers.GlmOcrForConditionalGeneration) (model GlmOcrConfig)
- **glmga** -- [Glm46VForConditionalGeneration](/docs/transformers/main/id/model_doc/glm46v#transformers.Glm46VForConditionalGeneration) (model GlmgaConfig)
- **got_ocr2** -- [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (model GotOcr2Config)
- **granite4_vision** -- [Granite4VisionForConditionalGeneration](/docs/transformers/main/id/model_doc/granite4_vision#transformers.Granite4VisionForConditionalGeneration) (model Granite4VisionConfig)
- **hunyuan_vl** -- [HunYuanVLForConditionalGeneration](/docs/transformers/main/id/model_doc/hunyuan_vl#transformers.HunYuanVLForConditionalGeneration) (model HunYuanVLConfig)
- **hyperclovax_vision_v2** -- [HyperCLOVAXVisionV2ForConditionalGeneration](/docs/transformers/main/id/model_doc/hyperclovax_vision_v2#transformers.HyperCLOVAXVisionV2ForConditionalGeneration) (model HyperCLOVAXVisionV2Config)
- **idefics** -- [IdeficsForVisionText2Text](/docs/transformers/main/id/model_doc/idefics#transformers.IdeficsForVisionText2Text) (model IdeficsConfig)
- **idefics2** -- [Idefics2ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics2#transformers.Idefics2ForConditionalGeneration) (model Idefics2Config)
- **idefics3** -- [Idefics3ForConditionalGeneration](/docs/transformers/main/id/model_doc/idefics3#transformers.Idefics3ForConditionalGeneration) (model Idefics3Config)
- **inkling_mm_model** -- [InklingForConditionalGeneration](/docs/transformers/main/id/model_doc/inkling#transformers.InklingForConditionalGeneration) (model InklingConfig)
- **instructblip** -- [InstructBlipForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblip#transformers.InstructBlipForConditionalGeneration) (model InstructBlipConfig)
- **instructblipvideo** -- [InstructBlipVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/instructblipvideo#transformers.InstructBlipVideoForConditionalGeneration) (model InstructBlipVideoConfig)
- **internvl** -- [InternVLForConditionalGeneration](/docs/transformers/main/id/model_doc/internvl#transformers.InternVLForConditionalGeneration) (model InternVLConfig)
- **janus** -- [JanusForConditionalGeneration](/docs/transformers/main/id/model_doc/janus#transformers.JanusForConditionalGeneration) (model JanusConfig)
- **kimi_k25** -- [Kimi_K25ForConditionalGeneration](/docs/transformers/main/id/model_doc/kimi_k25#transformers.Kimi_K25ForConditionalGeneration) (model Kimi_K25Config)
- **kosmos-2** -- [Kosmos2ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos-2#transformers.Kosmos2ForConditionalGeneration) (model Kosmos2Config)
- **kosmos-2.5** -- [Kosmos2_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/kosmos2_5#transformers.Kosmos2_5ForConditionalGeneration) (model Kosmos2_5Config)
- **lfm2_vl** -- [Lfm2VlForConditionalGeneration](/docs/transformers/main/id/model_doc/lfm2_vl#transformers.Lfm2VlForConditionalGeneration) (model Lfm2VlConfig)
- **lighton_ocr** -- [LightOnOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/lighton_ocr#transformers.LightOnOcrForConditionalGeneration) (model LightOnOcrConfig)
- **llama4** -- [Llama4ForConditionalGeneration](/docs/transformers/main/id/model_doc/llama4#transformers.Llama4ForConditionalGeneration) (model Llama4Config)
- **llava** -- [LlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/llava#transformers.LlavaForConditionalGeneration) (model LlavaConfig)
- **llava_next** -- [LlavaNextForConditionalGeneration](/docs/transformers/main/id/model_doc/granitevision#transformers.LlavaNextForConditionalGeneration) (model LlavaNextConfig)
- **llava_next_video** -- [LlavaNextVideoForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_next_video#transformers.LlavaNextVideoForConditionalGeneration) (model LlavaNextVideoConfig)
- **llava_onevision** -- [LlavaOnevisionForConditionalGeneration](/docs/transformers/main/id/model_doc/llava_onevision#transformers.LlavaOnevisionForConditionalGeneration) (model LlavaOnevisionConfig)
- **minicpm_ocr** -- [MiniCPMOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpm_ocr#transformers.MiniCPMOcrForConditionalGeneration) (model MiniCPMOcrConfig)
- **minicpmv4_6** -- [MiniCPMV4_6ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_6#transformers.MiniCPMV4_6ForConditionalGeneration) (model MiniCPMV4_6Config)
- **minicpmv4_7** -- [MiniCPMV4_7ForConditionalGeneration](/docs/transformers/main/id/model_doc/minicpmv4_7#transformers.MiniCPMV4_7ForConditionalGeneration) (model MiniCPMV4_7Config)
- **minimax_m3_vl** -- [MiniMaxM3SparseForConditionalGeneration](/docs/transformers/main/id/model_doc/minimax_m3_vl#transformers.MiniMaxM3SparseForConditionalGeneration) (model MiniMaxM3VLConfig)
- **mistral3** -- [Mistral3ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration) (model Mistral3Config)
- **mistral4** -- [Mistral4ForCausalLM](/docs/transformers/main/id/model_doc/mistral4#transformers.Mistral4ForCausalLM) (model Mistral4Config)
- **mistral_large4** -- [MistralLarge4ForConditionalGeneration](/docs/transformers/main/id/model_doc/mistral_large4#transformers.MistralLarge4ForConditionalGeneration) (model MistralLarge4Config)
- **mllama** -- [MllamaForConditionalGeneration](/docs/transformers/main/id/model_doc/mllama#transformers.MllamaForConditionalGeneration) (model MllamaConfig)
- **molmo2** -- [Molmo2ForConditionalGeneration](/docs/transformers/main/id/model_doc/molmo2#transformers.Molmo2ForConditionalGeneration) (model Molmo2Config)
- **muse_glimmer** -- [MuseGlimmerForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_glimmer#transformers.MuseGlimmerForConditionalGeneration) (model MuseGlimmerConfig)
- **muse_spark** -- [MuseSparkForConditionalGeneration](/docs/transformers/main/id/model_doc/muse_spark#transformers.MuseSparkForConditionalGeneration) (model MuseSparkConfig)
- **nemotron_h_omni** -- [NemotronH_Omni_Reasoning_V3](/docs/transformers/main/id/model_doc/nemotron_h_omni#transformers.NemotronH_Omni_Reasoning_V3) (model NemotronH_Omni_Reasoning_V3_Config)
- **ovis2** -- [Ovis2ForConditionalGeneration](/docs/transformers/main/id/model_doc/ovis2#transformers.Ovis2ForConditionalGeneration) (model Ovis2Config)
- **paddleocr_vl** -- [PaddleOCRVLForConditionalGeneration](/docs/transformers/main/id/model_doc/paddleocr_vl#transformers.PaddleOCRVLForConditionalGeneration) (model PaddleOCRVLConfig)
- **paligemma** -- [PaliGemmaForConditionalGeneration](/docs/transformers/main/id/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration) (model PaliGemmaConfig)
- **perception_lm** -- [PerceptionLMForConditionalGeneration](/docs/transformers/main/id/model_doc/perception_lm#transformers.PerceptionLMForConditionalGeneration) (model PerceptionLMConfig)
- **pi0** -- [PI0ForConditionalGeneration](/docs/transformers/main/id/model_doc/pi0#transformers.PI0ForConditionalGeneration) (model PI0Config)
- **pix2struct** -- [Pix2StructForConditionalGeneration](/docs/transformers/main/id/model_doc/pix2struct#transformers.Pix2StructForConditionalGeneration) (model Pix2StructConfig)
- **pp_chart2table** -- [GotOcr2ForConditionalGeneration](/docs/transformers/main/id/model_doc/got_ocr2#transformers.GotOcr2ForConditionalGeneration) (model PPChart2TableConfig)
- **pp_formulanet** -- [PPFormulaNetForConditionalGeneration](/docs/transformers/main/id/model_doc/pp_formulanet#transformers.PPFormulaNetForConditionalGeneration) (model PPFormulaNetConfig)
- **qianfan_ocr** -- [QianfanOCRForConditionalGeneration](/docs/transformers/main/id/model_doc/qianfan_ocr#transformers.QianfanOCRForConditionalGeneration) (model QianfanOCRConfig)
- **qwen2_5_omni_thinker** -- [Qwen2_5OmniThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_omni#transformers.Qwen2_5OmniThinkerForConditionalGeneration) (model Qwen2_5OmniThinkerConfig)
- **qwen2_5_vl** -- [Qwen2_5_VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_5_vl#transformers.Qwen2_5_VLForConditionalGeneration) (model Qwen2_5_VLConfig)
- **qwen2_vl** -- [Qwen2VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen2_vl#transformers.Qwen2VLForConditionalGeneration) (model Qwen2VLConfig)
- **qwen3_5** -- [Qwen3_5ForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5#transformers.Qwen3_5ForConditionalGeneration) (model Qwen3_5Config)
- **qwen3_5_moe** -- [Qwen3_5MoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_5_moe#transformers.Qwen3_5MoeForConditionalGeneration) (model Qwen3_5MoeConfig)
- **qwen3_omni_moe_thinker** -- [Qwen3OmniMoeThinkerForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_omni_moe#transformers.Qwen3OmniMoeThinkerForConditionalGeneration) (model Qwen3OmniMoeThinkerConfig)
- **qwen3_vl** -- [Qwen3VLForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl#transformers.Qwen3VLForConditionalGeneration) (model Qwen3VLConfig)
- **qwen3_vl_moe** -- [Qwen3VLMoeForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen3_vl_moe#transformers.Qwen3VLMoeForConditionalGeneration) (model Qwen3VLMoeConfig)
- **qwen4_exp** -- [Qwen4ExpForConditionalGeneration](/docs/transformers/main/id/model_doc/qwen4_exp#transformers.Qwen4ExpForConditionalGeneration) (model Qwen4ExpConfig)
- **shieldgemma2** -- [Gemma3ForConditionalGeneration](/docs/transformers/main/id/model_doc/gemma3#transformers.Gemma3ForConditionalGeneration) (model ShieldGemma2Config)
- **smolvlm** -- [SmolVLMForConditionalGeneration](/docs/transformers/main/id/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration) (model SmolVLMConfig)
- **step3p7** -- [Step3p7ForConditionalGeneration](/docs/transformers/main/id/model_doc/step3p7#transformers.Step3p7ForConditionalGeneration) (model Step3p7Config)
- **t5gemma2** -- [T5Gemma2ForConditionalGeneration](/docs/transformers/main/id/model_doc/t5gemma2#transformers.T5Gemma2ForConditionalGeneration) (model T5Gemma2Config)
- **udop** -- [UdopForConditionalGeneration](/docs/transformers/main/id/model_doc/udop#transformers.UdopForConditionalGeneration) (model UdopConfig)
- **unlimited_ocr** -- [UnlimitedOcrForConditionalGeneration](/docs/transformers/main/id/model_doc/unlimited_ocr#transformers.UnlimitedOcrForConditionalGeneration) (model UnlimitedOcrConfig)
- **video_llama_3** -- [VideoLlama3ForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llama_3#transformers.VideoLlama3ForConditionalGeneration) (model VideoLlama3Config)
- **video_llava** -- [VideoLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/video_llava#transformers.VideoLlavaForConditionalGeneration) (model VideoLlavaConfig)
- **vipllava** -- [VipLlavaForConditionalGeneration](/docs/transformers/main/id/model_doc/vipllava#transformers.VipLlavaForConditionalGeneration) (model VipLlavaConfig)
- **vision-encoder-decoder** -- [VisionEncoderDecoderModel](/docs/transformers/main/id/model_doc/vision-encoder-decoder#transformers.VisionEncoderDecoderModel) (model VisionEncoderDecoderConfig)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForImageTextToText

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForImageTextToText.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForImageTextToText.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```

## Deret Waktu

### AutoModelForTimeSeriesPrediction[[transformers.AutoModelForTimeSeriesPrediction]]

#### transformers.AutoModelForTimeSeriesPrediction[[transformers.AutoModelForTimeSeriesPrediction]]

```python
transformers.AutoModelForTimeSeriesPrediction(*args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/modeling_auto.py#L2450)

Ini adalah kelas model generik yang akan diinstansiasi sebagai salah satu kelas model dari pustaka (dengan kepala prediksi deret waktu) saat dibuat dengan metode kelas [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) atau metode kelas [from_config()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_config).

Kelas ini tidak dapat diinstansiasi secara langsung menggunakan `__init__()` (melempar error).

#### from_config[[transformers.AutoModelForTimeSeriesPrediction.from_config]]

```python
from_config(**kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L204)

**Parameter:**

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig)) : Kelas model yang akan diinstansiasi dipilih berdasarkan kelas konfigurasi:  - Kelas konfigurasi [TimesFm2_5Config](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5Config): [TimesFm2_5ModelForPrediction](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5ModelForPrediction) (model TimesFm2_5Config) - Kelas konfigurasi [TimesFmConfig](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmConfig): [TimesFmModelForPrediction](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmModelForPrediction) (model TimesFmConfig)

attn_implementation (`str`, *opsional*) : Implementasi attention yang akan digunakan dalam model (jika relevan). Bisa berupa salah satu dari - `"eager"` (implementasi manual dari attention) - `"sdpa"` (menggunakan [`F.scaled_dot_product_attention`](https://pytorch.org/docs/master/generated/torch.nn.functional.scaled_dot_product_attention.html)) - `"flash_attention_2"` (menggunakan [Dao-AILab/flash-attention](https://github.com/Dao-AILab/flash-attention)) - `"flash_attention_3"` (menggunakan [Dao-AILab/flash-attention/hopper](https://github.com/Dao-AILab/flash-attention/tree/main/hopper)) - `"flash_attention_4"` (menggunakan [Dao-AILab/flash-attention/flash_attn/cute](https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/cute)). Secara default, jika tersedia, SDPA akan digunakan untuk torch>=2.1.1. Default lainnya adalah implementasi manual `"eager"`.

Membuat instans salah satu kelas model dari pustaka (dengan kepala prediksi deret waktu) dari sebuah konfigurasi.

Catatan: Memuat model dari file konfigurasinya **tidak** memuat bobot model. Hal ini hanya memengaruhi konfigurasi model. Gunakan [from_pretrained()](/docs/transformers/main/id/model_doc/auto#transformers.AutoModel.from_pretrained) untuk memuat bobot model.

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTimeSeriesPrediction

>>> # Download configuration from huggingface.co and cache.
>>> config = AutoConfig.from_pretrained("google-bert/bert-base-cased")
>>> model = AutoModelForTimeSeriesPrediction.from_config(config)
```

#### from_pretrained[[transformers.AutoModelForTimeSeriesPrediction.from_pretrained]]

```python
from_pretrained(*model_args, **kwargs)
```

[Sumber](https://github.com/huggingface/transformers/blob/main/src/transformers/models/auto/auto_factory.py#L260)

**Parameter:**

pretrained_model_name_or_path (`str` atau `os.PathLike`) : Bisa berupa:  - String, *model id* dari model pretrained yang dihosting di dalam repo model di huggingface.co. - Path ke *directory* yang berisi bobot model yang disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained), misalnya `./my_model_directory/`.

model_args (argumen posisi tambahan, *opsional*) : Akan diteruskan ke metode `__init__()` model yang mendasarinya.

config ([PreTrainedConfig](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig), *opsional*) : Konfigurasi untuk model yang akan digunakan sebagai pengganti konfigurasi yang dimuat secara otomatis. Konfigurasi dapat dimuat secara otomatis ketika: - Model adalah model yang disediakan oleh pustaka (dimuat dengan string *model id* dari model pretrained). - Model disimpan menggunakan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan dimuat ulang dengan menyediakan direktori penyimpanan. - Model dimuat dengan menyediakan direktori lokal sebagai `pretrained_model_name_or_path` dan file konfigurasi JSON bernama *config.json* ditemukan di direktori tersebut.

state_dict (*dict[str, torch.Tensor]*, *opsional*) : Kamus status yang akan digunakan sebagai pengganti kamus status yang dimuat dari file bobot yang disimpan. Opsi ini dapat digunakan jika Anda ingin membuat model dari konfigurasi pralatih tetapi memuat bobot Anda sendiri. Namun, dalam kasus ini, Anda sebaiknya memeriksa apakah penggunaan [save_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.save_pretrained) dan [from_pretrained()](/docs/transformers/main/id/main_classes/model#transformers.PreTrainedModel.from_pretrained) bukan merupakan opsi yang lebih sederhana.

cache_dir (`str` atau `os.PathLike`, *opsional*) : Path ke direktori tempat konfigurasi model pretrained yang diunduh harus di-cache jika cache standar tidak digunakan.

force_download (`bool`, *opsional*, default `False`) : Apakah akan memaksa (mengunduh ulang) bobot model dan file konfigurasi, menimpa versi yang sudah di-cache jika ada.

proxies (`dict[str, str]`, *opsional*) : Kamus server proxy yang digunakan berdasarkan protokol atau endpoint, misalnya `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}`. Proxy digunakan pada setiap permintaan.

output_loading_info(`bool`, *opsional*, default `False`) : Apakah akan mengembalikan kamus yang berisi kunci yang hilang, kunci yang tidak terduga, dan pesan kesalahan.

local_files_only(`bool`, *opsional*, default `False`) : Apakah hanya melihat file lokal (misalnya, tidak mencoba mengunduh model).

revision (`str`, *opsional*, defaultnya `"main"`) : Versi model spesifik yang akan digunakan. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

trust_remote_code (`bool`, *opsional*, default `False`) : Apakah model kustom yang didefinisikan di Hub dalam file modeling mereka sendiri diizinkan atau tidak. Opsi ini hanya harus disetel ke `True` untuk repositori yang Anda percaya dan yang kodenya telah Anda baca, karena opsi ini akan mengeksekusi kode yang ada di Hub pada mesin lokal Anda.

code_revision (`str`, *opsional*, defaultnya `"main"`) : Revisi spesifik yang digunakan untuk kode di Hub, jika kode berada di repositori yang berbeda dari bagian lain model. Ini bisa berupa nama branch, nama tag, atau id commit, karena kami menggunakan sistem berbasis git untuk menyimpan model dan artefak lainnya di huggingface.co, sehingga `revision` dapat berupa identifikasi apa pun yang diizinkan oleh git.

kwargs (argumen kata kunci tambahan, *opsional*) : Dapat digunakan untuk memperbarui objek konfigurasi (setelah dimuat) dan menginisiasi model (misalnya, `output_attentions=True`). Perilakunya berbeda tergantung pada apakah `config` disediakan atau dimuat secara otomatis: - Jika konfigurasi disediakan dengan `config`, `**kwargs` akan langsung diteruskan ke metode `__init__` model yang mendasarinya (kami asumsikan semua pembaruan relevan pada konfigurasi sudah dilakukan) - Jika konfigurasi tidak disediakan, `kwargs` akan terlebih dahulu diteruskan ke fungsi inisialisasi kelas konfigurasi ([from_pretrained()](/docs/transformers/main/id/main_classes/configuration#transformers.PreTrainedConfig.from_pretrained)). Setiap kunci dari `kwargs` yang sesuai dengan atribut konfigurasi akan digunakan untuk menimpa atribut tersebut dengan nilai `kwargs` yang diberikan. Kunci sisa yang tidak sesuai dengan atribut konfigurasi apa pun akan diteruskan ke fungsi `__init__` model yang mendasarinya.

Instansiasi salah satu kelas model dari pustaka (dengan kepala prediksi deret waktu) dari model pralatih.

Kelas model yang akan diinstansiasi dipilih berdasarkan properti `model_type` dari objek config (baik yang dilewatkan sebagai argumen maupun dimuat dari `pretrained_model_name_or_path` jika memungkinkan), atau jika properti tersebut tidak ada, dengan menggunakan pencocokan pola pada `pretrained_model_name_or_path`:

- **timesfm** -- [TimesFmModelForPrediction](/docs/transformers/main/id/model_doc/timesfm#transformers.TimesFmModelForPrediction) (model TimesFmConfig)
- **timesfm2_5** -- [TimesFm2_5ModelForPrediction](/docs/transformers/main/id/model_doc/timesfm2_5#transformers.TimesFm2_5ModelForPrediction) (model TimesFm2_5Config)

Model secara default diatur dalam mode evaluasi menggunakan `model.eval()` (misalnya, modul dropout dinonaktifkan). Untuk melatih model, Anda harus terlebih dahulu mengembalikannya ke mode pelatihan dengan `model.train()`

Contoh:

```python
>>> from transformers import AutoConfig, AutoModelForTimeSeriesPrediction

>>> # Download model and configuration from huggingface.co and cache.
>>> model = AutoModelForTimeSeriesPrediction.from_pretrained("google-bert/bert-base-cased")

>>> # Update configuration during loading
>>> model = AutoModelForTimeSeriesPrediction.from_pretrained("google-bert/bert-base-cased", output_attentions=True)
>>> model.config.output_attentions
True
```