Instructions to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF 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 tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
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 tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
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 tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
Use Docker
docker model run hf.co/tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
- LM Studio
- Jan
- Ollama
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Ollama:
ollama run hf.co/tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
- Unsloth Studio
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF to start chatting
- Pi
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- OpenClaw new
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
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 "tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M" \ --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"
- Docker Model Runner
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Docker Model Runner:
docker model run hf.co/tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
- Lemonade
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.jp-jitsumu-calc-qwen3-4b-GGUF-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
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 tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF:Q4_K_M
Run Hermes
hermes
- Atomic Chat
jp-jitsumu-calc-qwen3-4b-GGUF
jp-jitsumu-calc-qwen3-4b(日本語業務の多段計算特化・期日計算/複数書類突合)のGGUF版です。llama.cppでWindows/Linux/MacのCPU・GPUで動作します。
ファイルと実測
| ファイル | サイズ | 期日計算(dev) | 書類突合(dev) | 備考 |
|---|---|---|---|---|
| Q8_0 | 4.0GB | 1.000 | 1.000 | 推奨。MLXアダプタ版と同値 |
| Q4_K_M | 2.3GB | 0.875 | 0.975 | 省メモリ。軽い劣化あり |
| f16 | 7.5GB | — | — | 参照用 |
検証セット(各40問・学習と分離)での正答率。本体repoのテスト値(期日計算0.925・書類突合0.975、ベース比+0.575/+0.425)はMLXアダプタ構成での測定です。
使い方
# llama.cpp(brew install llama.cpp 等で導入)
llama-server -m jp-jitsumu-calc-qwen3-4b-Q8_0.gguf --port 8080 --jinja --reasoning-format none
--reasoning-format none を推奨します。 既定のパーサだと本モデルの計算過程がreasoning_content側に分離され、contentが空になる場合があります。出力冒頭に迷いトークン(<think>等)が現れることがありますが、最終行のJSON({"answer": ...})をパースする運用で機械処理できます。
プロンプトは学習時の定型をそのまま使ってください(形式を変えると精度が下がります):
あなたは日本企業の経理・総務担当者です。以下の指示を読み、規程に従って正確に計算し、答えを求めてください。
# 指示
(ここに期日計算・書類突合の問題文)
# 出力形式
最終的な答えのみを、次のJSON1行で出力してください(説明文・単位・カンマは付けない)。
{"answer": <答え>}
- 日付を問われた場合: "answer" は "YYYY-MM-DD" 形式の文字列にすること(例: {"answer": "2026-03-02"})
- 金額・個数・数値を問われた場合: "answer" は数値のみにすること(カンマ・「円」「箱」等の単位を付けない。例: {"answer": 35000})
データと帰属
- 学習データ: プログラム生成した業務計算問題(正解既知)に対する DeepSeek-V4-Flash の検証済み解答軌跡(DeepSeek Open Platform利用規約は蒸留を明示的に許可)
- ベース: Qwen/Qwen3-4B-Instruct-2507(Apache-2.0)
Developed by tokimoa — データを外に出さないAI活用を支援しています
- Downloads last month
- 75
4-bit
8-bit
16-bit
Model tree for tokimoa/jp-jitsumu-calc-qwen3-4b-GGUF
Base model
Qwen/Qwen3-4B-Instruct-2507