Instructions to use kk0518/Nagaki-4B-FFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use kk0518/Nagaki-4B-FFT with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kk0518/Nagaki-4B-FFT:F16
# Run inference directly in the terminal:
llama cli -hf kk0518/Nagaki-4B-FFT:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kk0518/Nagaki-4B-FFT:F16
# Run inference directly in the terminal:
llama cli -hf kk0518/Nagaki-4B-FFT:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf kk0518/Nagaki-4B-FFT:F16
# Run inference directly in the terminal:
./llama-cli -hf kk0518/Nagaki-4B-FFT:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf kk0518/Nagaki-4B-FFT:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf kk0518/Nagaki-4B-FFT:F16

Use Docker

docker model run hf.co/kk0518/Nagaki-4B-FFT:F16

LM Studio
Jan
Ollama
How to use kk0518/Nagaki-4B-FFT with Ollama:
```
ollama run hf.co/kk0518/Nagaki-4B-FFT:F16
```
Unsloth Desktop

How to use kk0518/Nagaki-4B-FFT with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kk0518/Nagaki-4B-FFT:F16

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "kk0518/Nagaki-4B-FFT:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use kk0518/Nagaki-4B-FFT with Docker Model Runner:
```
docker model run hf.co/kk0518/Nagaki-4B-FFT:F16
```

Lemonade

How to use kk0518/Nagaki-4B-FFT with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kk0518/Nagaki-4B-FFT:F16

Run and chat with the model

lemonade run user.Nagaki-4B-FFT-F16

List all available models

lemonade list

Hermes Agent

How to use kk0518/Nagaki-4B-FFT with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kk0518/Nagaki-4B-FFT:F16

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default kk0518/Nagaki-4B-FFT:F16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use kk0518/Nagaki-4B-FFT with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kk0518/Nagaki-4B-FFT:F16

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "kk0518/Nagaki-4B-FFT:F16" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Nagaki-4B-FFT-v2 (CreateLLM_JP)

📝 Dataset Metadata

本モデルは TCreateLLM_JP プロジェクトの一環として作成した、以下の合計 4,261,033 レコード の高品質な日本語データセットの一部を用いてフルファインチューニングされました。 TCreateLLM_JPプロジェクトは、今後名称を変更する可能性がありますのでご確認ください。

Hugging Face Space >>> https://huggingface.co/spaces/kk0518/nagaki_4b

Support & Donation / 開発支援のお願い

nagakiシリーズをご活用いただきありがとうございます。 nagaki_4b_fft では、おかげさまで1,300ダウンロードを突破する反響をいただきました。

本プロジェクトは個人での研究・開発となっており、より高性能なモデルの開発には多大な計算リソースが必要となります。もし本モデルがお役に立ちましたら、継続的な開発支援をいただけますと幸いです。

OFUSE（投げ銭）: https://ofuse.me/a753ea67
フィードバック: 使用感や改善点など、DiscussionやSNSで教えていただけると励みになります！

📊 Statistics

Wikipedia (Japanese): 2,073,483 records
Aozora Bunko: 1,008,756 records
Nemotron-Personas-Japan: 1,000,000 records
Tanaka Corpus: 147,876 records
Law QA (Digital Agency): 140 records デジタル庁"日本の法令に関する多肢選択式QAデータセット" ( https://github.com/digital-go-jp/lawqa_jp )を加工し作成
Other specialized data: Approx. 30,000 records (Weather, Honorifics, Trends, etc.)

Downloads last month: 68

GGUF

Model size

4B params

Architecture

qwen35

Hardware compatibility

4-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for kk0518/Nagaki-4B-FFT

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Quantized

(514)

this model

kk0518
/

Nagaki-4B-FFT

Nagaki-4B-FFT-v2 (CreateLLM_JP)

📝 Dataset Metadata

Support & Donation / 開発支援のお願い

📊 Statistics

Model tree for kk0518/Nagaki-4B-FFT

Datasets used to train kk0518/Nagaki-4B-FFT

Space using kk0518/Nagaki-4B-FFT 1