AI and models · Independent · 22 Oct 2026

unslothai/unsloth

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

← AI and models
Unsloth: local AI runner and trainer. Run and fine-tune models on your hardware.

In simple English

This tool lets you run and train large artificial intelligence models on your own computer. You can fine-tune models to fit your specific tasks without needing expensive cloud servers. It comes with a desktop app and a web interface so you do not have to write code to use it. You can also connect your local models to coding assistants and other developer tools. The core software is free to use, though some parts use copyleft rules.

What it is

Unsloth is a desktop application, web user interface, and Python library designed for running, fine-tuning, and deploying artificial intelligence models locally. It supports major open model families including Qwen, DeepSeek, Gemma, and FLUX.

What it is good for

Solo builders who want to run private artificial intelligence models locally, fine-tune open-weights models on custom datasets, or connect local models to coding tools like Claude Code without paying for API subscriptions.

Stars78k
LicenceApache-2.0
Latestv0.1.909-beta (2026-10-19)checked 22 Oct 2026
Last code change2026-10-22
Written inPython
Setup effort
3 / 5
Payoff
5 / 5
Good
  • Runs and trains models with significantly less memory than standard tools.
  • Provides a native desktop app and web interface alongside a Python library.
  • Connects easily to popular coding agents and supplies an OpenAI-compatible local API.
Watch for
  • Requires a powerful graphics card (GPU) with plenty of VRAM to train or run large models.
  • Dual licensing means some components fall under AGPL-3.0 restrictions.
  • High configuration complexity if you attempt manual setup across different hardware accelerators.
Replaces

Manual PyTorch training scripts and cloud-based fine-tuning services

Works well with

Claude Code

Similar repositories
  • ggml-org/llama.cpp132k stars

    Runs large language models locally from the command line with low memory usage, but lacks a built-in training studio.

  • oobabooga/textgen48k stars

    Provides a local web interface for running and chatting with language models, though it has less focus on accelerated fine-tuning.

  • ollama/ollama184k stars

    Runs local models easily via a simple command-line tool, but does not offer built-in model training or fine-tuning workflows.

Also reviewed on Repo Radar
  • ggml-org/llama.cpp132k stars

    A C and C++ library and command-line tool for running language models locally.

  • EleutherAI/lm-evaluation-harness14k stars

    Runs standard question sets against a language model and reports the scores.

  • mudler/LocalAI50k stars

    Serves text, vision, speech, image and video models from one self-hosted service that imitates the paid interfaces.

Install
curl -fsSL https://unsloth.ai/install.sh | sh
unsloth studio

Independent: reviewed for Repo Radar outside the daily editions. Every independent repository

On Stack Radar: AI & Automation › AI Models & Inference › Model Fine-Tuning.

Get the next edition in your inbox

A dozen repositories, opened and checked. The licence read, the last release dated, and the ones that did not make it named with the reason. It is the half most lists leave out.

No tracking pixels. One click to leave. The archive stays free either way.

We use your address to send the edition and nothing else. Confirm by email, leave in one click. How we handle it.