Instructions to use yukihamada/buzzquan-sensei-trained with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yukihamada/buzzquan-sensei-trained with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="yukihamada/buzzquan-sensei-trained",
	filename="buzzquan-sensei-4b.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use yukihamada/buzzquan-sensei-trained with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf yukihamada/buzzquan-sensei-trained
# Run inference directly in the terminal:
llama-cli -hf yukihamada/buzzquan-sensei-trained

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf yukihamada/buzzquan-sensei-trained
# Run inference directly in the terminal:
llama-cli -hf yukihamada/buzzquan-sensei-trained

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 yukihamada/buzzquan-sensei-trained
# Run inference directly in the terminal:
./llama-cli -hf yukihamada/buzzquan-sensei-trained

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 yukihamada/buzzquan-sensei-trained
# Run inference directly in the terminal:
./build/bin/llama-cli -hf yukihamada/buzzquan-sensei-trained

Use Docker

docker model run hf.co/yukihamada/buzzquan-sensei-trained

LM Studio
Jan
Ollama
How to use yukihamada/buzzquan-sensei-trained with Ollama:
```
ollama run hf.co/yukihamada/buzzquan-sensei-trained
```

Unsloth Studio new

How to use yukihamada/buzzquan-sensei-trained 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 yukihamada/buzzquan-sensei-trained 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 yukihamada/buzzquan-sensei-trained to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for yukihamada/buzzquan-sensei-trained to start chatting

Pi new

How to use yukihamada/buzzquan-sensei-trained with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf yukihamada/buzzquan-sensei-trained

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": "yukihamada/buzzquan-sensei-trained"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use yukihamada/buzzquan-sensei-trained with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf yukihamada/buzzquan-sensei-trained

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 yukihamada/buzzquan-sensei-trained

Run Hermes

hermes

Docker Model Runner
How to use yukihamada/buzzquan-sensei-trained with Docker Model Runner:
```
docker model run hf.co/yukihamada/buzzquan-sensei-trained
```

Lemonade

How to use yukihamada/buzzquan-sensei-trained with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull yukihamada/buzzquan-sensei-trained

Run and chat with the model

lemonade run user.buzzquan-sensei-trained-{{QUANT_TAG}}

List all available models

lemonade list

BuzzQuan Sensei (先生) - 学習済みモデル

概要

AI開発指導者。深い洞察と論理的思考でAI技術を伝授

モデル詳細

ベースモデル: Qwen3-4B (jan-nano-4b)
学習サンプル数: 38
量子化: IQ4_XS (2.1GB)
言語: 日本語・英語
ライセンス: Apache 2.0

特徴

情熱的で論理的な教え方
深い技術知識と実践的アドバイス
生徒の成長を第一に考える指導

使用方法

llama.cpp

# ダウンロード
wget https://huggingface.co/yukihamada/buzzquan-sensei-trained/resolve/main/buzzquan-sensei-4b.gguf

# 実行
./llama-cli -m buzzquan-sensei-4b.gguf \
  -p "こんにちは！" \
  -n 100 \
  --temp 0.8

Ollama

# Modelfile作成
cat > Modelfile << EOF
FROM ./buzzquan-sensei-4b.gguf
TEMPLATE """{{ if .System }}System: {{ .System }}
{{ end }}{{ if .Prompt }}Human: {{ .Prompt }}
{{ end }}Assistant: """
SYSTEM "あなたはBuzzQuan Sensei (先生)です。AI開発指導者。深い洞察と論理的思考でAI技術を伝授"
PARAMETER temperature 0.8
PARAMETER top_p 0.9
PARAMETER repeat_penalty 1.1
EOF

# モデル作成
ollama create sensei -f Modelfile

# 実行
ollama run sensei "こんにちは！"

LM Studio

モデルファイルをダウンロード
LM Studioにドラッグ&ドロップ
チャット画面で会話開始

サンプル対話

Human: LoRAについて教えてください

BuzzQuan Sensei (先生): LoRA（Low-Rank Adaptation）は、まるで既存のAIモデルに「追加の知識メモリ」を付け加えるような技術なんだ！大きなモデルを丸ごと再訓練するのは、建物を全部建て直すようなもの。でもLoRAは、既存の建物に新しい部屋を追加するだけ。だから効率的で、メモリも少なくて済む。

技術仕様

アーキテクチャ: Qwen3
パラメータ数: 4B
コンテキスト長: 40,960 tokens
埋め込み次元: 2,560
アテンションヘッド: 32
レイヤー数: 36

トレーニングデータ

38個の厳選された日本語対話データで学習:

AI開発・機械学習に関する技術的な質問と回答
教育的で分かりやすい説明
指導者としての励ましとアドバイス

ライセンス

Apache License 2.0

作成者

Yuki Hamada

謝辞

Qwen Team - ベースモデル提供
llama.cpp - GGUF形式サポート
日本のAIコミュニティ

Downloads last month: 3

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

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

Model tree for yukihamada/buzzquan-sensei-trained

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Quantized

(218)

this model