Text Generation
GGUF
Safetensors
Japanese
English
lfm2
liquid
lfm
quantization
business
expert
conversational
Instructions to use ryukin164/LFM2.5-1.2B-Q4-JP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use ryukin164/LFM2.5-1.2B-Q4-JP with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="ryukin164/LFM2.5-1.2B-Q4-JP", filename="LFM-Business-Perfect-Q4.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": "What is the capital of France?" } ] ) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use ryukin164/LFM2.5-1.2B-Q4-JP with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf ryukin164/LFM2.5-1.2B-Q4-JP # Run inference directly in the terminal: llama-cli -hf ryukin164/LFM2.5-1.2B-Q4-JP
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf ryukin164/LFM2.5-1.2B-Q4-JP # Run inference directly in the terminal: llama-cli -hf ryukin164/LFM2.5-1.2B-Q4-JP
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 ryukin164/LFM2.5-1.2B-Q4-JP # Run inference directly in the terminal: ./llama-cli -hf ryukin164/LFM2.5-1.2B-Q4-JP
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 ryukin164/LFM2.5-1.2B-Q4-JP # Run inference directly in the terminal: ./build/bin/llama-cli -hf ryukin164/LFM2.5-1.2B-Q4-JP
Use Docker
docker model run hf.co/ryukin164/LFM2.5-1.2B-Q4-JP
- LM Studio
- Jan
- vLLM
How to use ryukin164/LFM2.5-1.2B-Q4-JP with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ryukin164/LFM2.5-1.2B-Q4-JP" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ryukin164/LFM2.5-1.2B-Q4-JP", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/ryukin164/LFM2.5-1.2B-Q4-JP
- Ollama
How to use ryukin164/LFM2.5-1.2B-Q4-JP with Ollama:
ollama run hf.co/ryukin164/LFM2.5-1.2B-Q4-JP
- Unsloth Studio
How to use ryukin164/LFM2.5-1.2B-Q4-JP 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 ryukin164/LFM2.5-1.2B-Q4-JP 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 ryukin164/LFM2.5-1.2B-Q4-JP to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for ryukin164/LFM2.5-1.2B-Q4-JP to start chatting
- Pi
How to use ryukin164/LFM2.5-1.2B-Q4-JP with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf ryukin164/LFM2.5-1.2B-Q4-JP
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": "ryukin164/LFM2.5-1.2B-Q4-JP" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use ryukin164/LFM2.5-1.2B-Q4-JP with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf ryukin164/LFM2.5-1.2B-Q4-JP
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 ryukin164/LFM2.5-1.2B-Q4-JP
Run Hermes
hermes
- Atomic Chat new
- Docker Model Runner
How to use ryukin164/LFM2.5-1.2B-Q4-JP with Docker Model Runner:
docker model run hf.co/ryukin164/LFM2.5-1.2B-Q4-JP
- Lemonade
How to use ryukin164/LFM2.5-1.2B-Q4-JP with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull ryukin164/LFM2.5-1.2B-Q4-JP
Run and chat with the model
lemonade run user.LFM2.5-1.2B-Q4-JP-{{QUANT_TAG}}List all available models
lemonade list
LFM 2.5 1.2B ビジネス専門家 (Q4_K_M GGUF)
📌 モデル概要
本プロジェクトは、LiquidAI/LFM2.5-1.2B-JP をベースとした、ビジネスシーン、専門的なコンサルティング、および論理的推論に特化した量子化済みモデルです。
最新の llama.cpp を使用し、Q4_K_M 形式で量子化を行いました。731MB という極めて軽量なサイズながら、高い推論能力を維持しています。
- アーキテクチャ: LFM (Liquid Foundation Model) - 線形回帰と畳み込みを組み合わせた非 Transformer 構造。
- パラメータ数: 1.2B
- 量子化形式: GGUF (Q4_K_M)
- ファイルサイズ: 731 MB
- 主な用途: モバイルデバイスでの実行、低スペックサーバー、ビジネス対話エージェント。
🚀 使い方
1. llama.cpp で実行
./llama-cli -m LFM-Business-Perfect-Q4.gguf -n 512 --prompt "<|im_start|>user\n新規事業のキャッシュフローを最適化する方法を教えてください。<|im_end|>\n<|im_start|>assistant\n"
2. Python (llama-cpp-python) で実行Pythonfrom llama_cpp import Llama
llm = Llama(model_path="./LFM-Business-Perfect-Q4.gguf", n_ctx=2048)
output = llm("<|im_start|>user\nビジネスプランの添削をお願いします。<|im_end|>\n<|im_start|>assistant\n", max_tokens=512)
print(output["choices"][0]["text"])
🛠 量子化の詳細Q4_K_M 量子化により、モデルの知能指数を最大限に保持しつつ、メモリ消費を大幅に削減しました。1.2B クラスのモデルにおいて、CPU 環境での実行に最も適したバランスです。項目詳細オリジナルサイズ~2.5 GB量子化後サイズ731 MB推奨 RAM2GB 以上⚠️ 免責事項このモデルは学習および研究目的で公開されています。生成される回答はアルゴリズムによるものであり、正確性や法的効力を保証するものではありません。実際のビジネス判断に際しては、専門家にご相談ください。🤝 謝辞優れたベースモデルをオープンソースとして公開してくださった Liquid AI チームに深く感謝いたします。
- Downloads last month
- 179