Instructions to use tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF with Ollama:
```
ollama run hf.co/tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
```

Unsloth Studio

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF to start chatting

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF with Docker Model Runner:
```
docker model run hf.co/tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M
```

Lemonade

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tokimoa/jp-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.jp-gijiroku-qwen3-4b-v2-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use tokimoa/jp-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-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-gijiroku-qwen3-4b-v2-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

tokimoa commited on about 7 hours ago

Commit

fe2dbc2

verified ·

1 Parent(s): 5318ce5

Upload folder using huggingface_hub

Browse files

Files changed (5) hide show

.gitattributes +3 -0
README.md +53 -0
jp-gijiroku-qwen3-4b-v2-Q4_K_M.gguf +3 -0
jp-gijiroku-qwen3-4b-v2-Q8_0.gguf +3 -0
jp-gijiroku-qwen3-4b-v2-f16.gguf +3 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+jp-gijiroku-qwen3-4b-v2-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+jp-gijiroku-qwen3-4b-v2-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+jp-gijiroku-qwen3-4b-v2-f16.gguf filter=lfs diff=lfs merge=lfs -text

README.md ADDED Viewed

	@@ -0,0 +1,53 @@

+---
+license: apache-2.0
+base_model: Qwen/Qwen3-4B-Instruct-2507
+language:
+  - ja
+tags:
+  - gguf
+  - llama.cpp
+  - japanese
+  - meeting-minutes
+  - gijiroku
+---
+# jp-gijiroku-qwen3-4b-v2-GGUF
+日本語議事録抽出モデル [jp-gijiroku-qwen3-4b-v2](https://huggingface.co/tokimoa/jp-gijiroku-qwen3-4b-v2) のGGUF版です。[初代](https://huggingface.co/tokimoa/jp-gijiroku-qwen3-4b-GGUF)の後継で、**実際の60分Zoom会議の録音→話者分離→文字起こし→抽出パイプラインで見つけた弱点**を標的に継続学習しました。llama.cppでWindows/Linux/MacのCPU・GPUで動きます。
+## v2で直したこと（すべて実会議で確認）
+1. **日付の捏造をしない** — 本文に日付がなければ `date: "なし"`（初代は不定と明記していた点の解消）
+2. **完了報告を決定事項に混ぜない** — 「デプロイは終わってます」は決定でも宿題でもない
+3. **期限の誤借用をしない** — 別の話の「今週中」を無関係なタスクに付けない
+4. **話者ラベル付き転写に適応** — `話者A: ...` 形式の生転写を直接入力できる
+5. **報告文の中のコミットを宿題として拾う** — 「デプロイできたら結果を共有しておきます」→ todo（担当=発言者）
+## ファイルと実測（ASR劣化転写20件・全項目正解率）
+| ファイル | サイズ | 全項目正解 | 備考 |
+|---|---|---|---|
+| **Q8_0** | 4.0GB | **15/20** | 推奨 |
+| Q4_K_M | 2.3GB | 17/20 | 省メモリ |
+| f16 | 7.5GB | — | 参照用 |
+参考: 初代は同条件でQ8_0 16/20・Q4_K_M 19/20（n=20の揺らぎ圏・同帯域）。v2の価値は上の5項目=実会議で起きる系統的ミスの解消です。整った合成val（50件）は39/50で初代38/50と同等、弱点標的val（60件）は54/60です。
+## 使い方
+```bash
+llama-server -m jp-gijiroku-qwen3-4b-v2-Q8_0.gguf --port 8080 --jinja --reasoning-format none -c 8192
+```
+- **`--reasoning-format none` を推奨します。** 既定パーサだと判断過程がreasoning_content側に分離されcontentが空になる場合があります
+- 出力は「判断過程 → 最終行にJSON」です。**最終行をパース**してください（スキーマ: `{"date", "attendees", "decisions", "todos": [{"task","owner","due"}], "next_meeting"}`）
+- プロンプトは定型文「次の議事録から出席者・決定事項・宿題（担当と期限）・次回予定を抽出してJSONで出力してください。」+ 本文をそのまま使ってください（学習分布の形式）
+- **長時間会議（30分超）は10分前後のチャンクに分けて抽出→統合**を推奨します。学習分布（〜2千トークン）に近い入力長のほうが網羅性が大きく上がります（60分実会議で一括=決定1・宿題1に対しチャンク統合=決定7・宿題6）
+## データと帰属
+- 学習データ: 合成会議（正解既知・ASR劣化/話者ラベル/コミットあり・なし対比込み）に対する [DeepSeek-V4-Flash](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731) の検証済み解答軌跡（利用規約は蒸留を明示的に許可）
+- ベース: [Qwen/Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507)（Apache-2.0）
+---
+Developed by [tokimoa](https://tokimoa.jp) — データを外に出さないAI活用を支援しています

jp-gijiroku-qwen3-4b-v2-Q4_K_M.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f3aa142c24d4a3f14fab5de87e16eaadfe2c536bc4cc8ef5b491ce571bd1446a
+size 2497280736

jp-gijiroku-qwen3-4b-v2-Q8_0.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:18fe2239544fa0c39c6490b89064a434883ede474abaf1a60b6ec90497cd7049
+size 4280405216

jp-gijiroku-qwen3-4b-v2-f16.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:07d9e5d0e2e90213484405b4c35edc4e1d977cc6ee45fa6c72b56ef9166ad2cd
+size 8051285216