Qwen3.8-27B-INT8
INT8 quantization of Qwen/Qwen3.8-27B, produced with llm-compressor.
Vision encoder, gated-delta attention (conv1d / linear_attn), lm_head, embeddings, and norm layers are kept in original precision.
- Downloads last month
- 297
Model tree for lokeshe09/Qwen3.8-27B-INT8
Base model
Qwen/Qwen3.8-27B