AI Knowledge HubLL
LLaMA-Factory
An all-in-one tool to fine-tune many language models with LoRA and more, plus a web UI
| Category | Models, APIs & Runtimes |
|---|---|
| Runtime Environment | Personal PC · VPS · VPS (high-end GPU) |
| Install Difficulty | Medium |
| Official Site | https://github.com/hiyouga/LLaMA-Factory |
| GitHub | https://github.com/hiyouga/LLaMA-Factory |
| Install & Usage Docs | https://llamafactory.readthedocs.io/ |
What is it?
Fine-tune 100+ models with LoRA, SFT, and more, and run it from a web UI too. The big advantage is being able to try training without writing code.
What can it do?
- Supports 100+ models
- LoRA, QLoRA, and SFT
- Built-in web UI
- Quantization, merging, and inference in one
Who is it for?
Beginners to fine-tuning, or anyone wanting to experiment with many models.
Pros
- Web UI lowers the barrier
- Unifies many models and techniques
- Plenty of guides, including Korean
Cautions
- Real training still needs a GPU
- Large-scale training is memory-bound
Hands-on Testing
No hands-on test notes yet. Only verified findings get added here.