Unsloth

Overview

Unsloth is an open-source framework for running and training LLMs. It lets you run and train AI models on your own local hardware via an open-source UI. Supports MacOS, Linux, Windows, NVIDIA, AMD, Intel and CPU setups.

Key Features

• 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.
• Remote & LAN: Access your local models from any device on LAN or remotely through secure Cloudflare HTTPS.
• Fine-tuning: Train LLMs, diffusion, TTS, and embedding models 2× faster with 70% less VRAM with no accuracy loss.
• Export & Deploy: Export or Deploy models with including GGUF, NVFP4, FP8 and more formats.

Use Cases

• Run and train models locally for personal or research use.
• Connect local models to coding agents like Claude Code and Codex.
• Use private web search and RAG for deep research.
• Fine-tune models for specific tasks with reduced VRAM usage.

Getting Started

• Download the native desktop app for macOS, Windows, or Linux from the Unsloth website or GitHub Releases.
• Alternatively, install manually: on macOS, Linux, or WSL run `curl -fsSL https://unsloth\.ai/install\.sh | sh`; on Windows PowerShell run `irm https://unsloth\.ai/install\.ps1 | iex`.
• A Docker image `unsloth/unsloth` is also available.

Deployment & Requirements

• Supports MacOS, Linux, Windows, NVIDIA, AMD, Intel and CPU setups.
• Supports Multi GPU setups, NVIDIA, AMD, Intel GPUs, CPUs and the Vulkan backend.
• For Docker, GPU access setup may be required on Linux.

Before You Adopt

• License: Apache-2.0. Review its terms before using, modifying, or distributing the project.