開発ツール
peft
Fine-tune huge models on small hardware with LoRA and other efficient methods
★ 21.8k スターApache-2.0
ガイドを見る →LlamaFactoryは、hiyougaが公開しているオープンソースの開発ツールプロジェクトです（GitHub: hiyouga/LlamaFactory）。GitHubスターは75,362、フォークは9,239で、Apache-2.0ライセンスの下で公開されています。GitHubでの最終更新は2026年9月28日です。
Unified, efficient fine-tuning of 100+ LLMs and VLMs with a CLI and Web UI — ACL 2024
Decide what you want from LlamaFactory and check the README and requirements to confirm it fits your setup and Python environment.
Clone the repository from GitHub (github.com/hiyouga/LlamaFactory) or use the documented install channel if you do not need the source.
Follow the README quickstart to install dependencies and set any required configuration or credentials.
Work through the README's first example end to end, then adapt it to your own project or data.
Fine-tune over 100 LLMs and vision-language models including Llama, Qwen, Gemma and DeepSeek families.
Supports pre-training, SFT, RLHF, LoRA, QLoRA and quantization workflows.
Llama Board GUI (Gradio) lets you configure and launch training without writing code.
Actively adds new cutting-edge models as they are released.
Ships curated datasets for instruction tuning and preference training.
Official Docker images, including Ascend NPU variants, for reproducible training.
Start by getting the source and entering the project directory.
git clone https://github.com/hiyouga/LlamaFactory.git
cd LlamaFactoryClone shallow and install in editable mode, per the README quickstart.
git clone --depth 1 https://github.com/hiyouga/LlamaFactory.git
cd LlamaFactory
pip install -e .Start Llama Board to configure your first fine-tune in the browser.
llamafactory-cli webui # or: uv run llamafactory-cli webuiIt is maintained by hiyouga and contributors on GitHub, where the repository github.com/hiyouga/LlamaFactory is the canonical home for code, issues and releases.
Yes — it is open source under the Apache-2.0 license. Read the license in the repository for what is permitted, including any conditions on redistribution or commercial use.
No API key needed — training is local on your hardware; you need a suitable GPU for most models.
Compared with Hugging Face TRL alone: LlamaFactory unifies many models, methods and datasets behind one CLI and Web UI instead of wiring libraries together yourself.
Clone github.com/hiyouga/LlamaFactory, follow the README quickstart to install it, and run the documented first example before adapting it to your own work.
Fine-tune huge models on small hardware with LoRA and other efficient methods
A streamlined framework for evaluating and benchmarking LLMs, VLMs and AIGC models.
A local-first database client with AI SQL help — 40+ databases, desktop, web, Docker and CLI
Fine-tune and deploy 600+ LLMs and 300+ multimodal models with PEFT or full-parameter training.
Trafilatura — a Python and CLI tool to crawl the web and extract clean text and metadata for corpora and RAG.
A Python metasearch library aggregating many web search services — plus CLI, API and MCP server
LibreAIカタログに掲載された開発ツールプロジェクトの中で最もスターの多いもの — LlamaFactoryが合わなかった場合に検討したい、現実的な候補リストです。
The open source coding agent
The model-definition framework behind modern AI — thousands of pretrained transformers for text, vision, and audio.
Anthropic's agentic coding tool that lives in your terminal and works across your whole codebase
Cross-platform desktop all-in-one manager for Claude Code, Codex, OpenCode and other coding agents
OpenAI's lightweight coding agent that runs locally in your terminal
カタログの中でLlamaFactoryとスター数が近い開発ツールプロジェクト — 前後の棚もそのまま見て回れます。
CLI proxy that cuts LLM token consumption 60-90% on common dev commands — one Rust binary
Compress tool outputs, logs and RAG chunks before they reach the LLM — same answers, fewer tokens
A free course to get into large language models, with roadmaps and runnable Colab notebooks
Spec-driven development (SDD) for AI coding assistants.
Adaptive web scraping framework in Python — from a single request to a full-scale crawl
Autonomous coding agent available as an SDK, IDE extension and CLI assistant