GH Repository · unslothai
unsloth
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.
- stars
- 76,189
- 30-day movement
- +24281/day
- Related entries
- 60
- Connections
- 1
agent-appPythonstable-diffusionimage-generationuiaiself-hostedfine-tuninggemmatext-to-speechopenaillamaunslothttsqwenpythonllmchatgptllmsreinforcement-learningdeepseekagent
Unsloth is a self-hosted local UI for running and training LLMs and diffusion models, written in Python and licensed under Apache-2.0. Its description lists support for GGUF and MLX formats and models such as Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, and FLUX.
You want a single local interface to run and fine-tune LLMs and diffusion models without relying on cloud services.
Use it to
- Run LLMs locally through a self-hosted UI
- Fine-tune models like Qwen, Gemma, or DeepSeek
- Generate images with diffusion models such as FLUX or Stable Diffusion
- Work with GGUF and MLX model formats
- Experiment with reinforcement learning training
For Developers and hobbyists self-hosting and fine-tuning models locally
- Role
- agent-app
- Language
- Python
- Licence
- Apache-2.0
- Forks
- 6,945
- Open issues
- 889
- Last push
- 2026-09-15
- Latest release
- July-2024 · 2024-07-03
topicsfine-tuningllmself-hosteduidiffusionpython