🦥Unsloth Docs

Unsloth is an open-source framework for running and training LLMs.

Unsloth lets you run and train AI models on your own local hardware via an open-source UI.

Our docs will guide you through running & training your own LLM locally.

Download Features GitHub

Decision model

Turn any model into a decision model.

Qwen3.8-27B

Run and train Qwen3.8-27B in Unsloth.

EmbeddingGemma 2

Google's new multimodal embedding mode.

Qwen-Image-2.1

Run the 7B text to image model locally!

Qwen3.8-Flash-Next

New multimodal MoE model with Qwen4 arch.

Meet Unsloth Desktop

The first desktop app to run and train models.

⚡ Quickstart

Unsloth supports MacOS, Linux, Windows, NVIDIA, AMD, Intel and CPU setups. See: Unsloth Requirements. Download the native desktop app for your operating system:

Download Unsloth Desktop

Or, install Unsloth Studio natively, our browser-based web UI:

Install Unsloth:

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

Use the same command to update.

Install Unsloth:

irm https://unsloth.ai/install.ps1 | iex

Use the same command to update.

The Unsloth Docker image unsloth/unsloth is available on Docker. Read guide.

Install Unsloth Desktop:

brew install --cask unsloth

Update Unsloth Desktop:

brew upgrade --cask unsloth

Install Unsloth Desktop:

winget install --id UnslothAI.Unsloth -e

Update Unsloth Desktop:

winget upgrade --id UnslothAI.Unsloth -e

Complete LLM Directory 🧬Fine-tuning Guide

🔮Models Unsloth API

👾 Unsloth Start

Unsloth Start lets you connect Claude Code, Codex and other agents to local models via the unsloth start command.

Start Unsloth, load a model, open your project folder, and then run:

unsloth start claude

Replace claude with any agent below:

Agent

Command

Claude Code

unsloth start claude

OpenAI Codex

unsloth start codex

Visual Studio Code

unsloth start vscode

DeepSeek Harness

unsloth start dsh

OpenCode

unsloth start opencode

Hermes Agent

unsloth start hermes

OpenClaw

unsloth start openclaw

🦥 Why Unsloth?

We directly collab with teams behind gpt-oss, Qwen3, Llama 4, Mistral, Gemma 1-3 and Phi-4, where we’ve fixed critical bugs that greatly improved model accuracy. Andrej Karpathy for example has praised our work.
Unsloth streamlines local training, inference, data, and deployment
Unsloth supports inference and training for 500+ models: vision, TTS, embedding, RL

⭐ Features

Unsloth lets you run and train models for text, audio, embedding, vision and more. Unsloth provides many key features for both inference and training:

Inference

Run and train LLMs, diffusion, embedding, audio models: Qwen3.8, Kimi K3, MiniMax-H3, Muse Glimmer, DeepSeek-V4, Gemma 4.
Agents & Tools: Use local models with Claude Code, Codex, and MCP, including tool calling and code execution.
Search & RAG: Use private and unlimited web search, deep research, auto-compaction (rolling context window) and RAG.
Image and video: Run and train image and video diffusion or multimodal models
Remote & LAN: Access your local models from any device on LAN or remotely through secure Cloudflare HTTPS.
Connect: Serve models through an OpenAI compatible API. Also connect your ChatGPT/Codex subscription and cloud providers

Train & Deploy

Fine-tuning: Train LLMs, diffusion, TTS, and embedding models 2× faster with 70% less VRAM with no accuracy loss
Complete support: Supports reinforcement learning, LoRA, QLoRA, full fine tuning, pretraining, RL, GRPO, DPO, and FP8.
Export & Deploy: Export or Deploy models with including GGUF, NVFP4, FP8 and more formats.
Datasets: Build datasets from PDFs, CSVs, DOCX files, and more with Data Recipes.