Unsloth Unsloth AI

ai deep-learning diffusion fine-tuning large-language-model llm machine-learning

Install Unsloth with WinGet

winget install --id=UnslothAI.Unsloth -e

Unsloth Desktop is an AI development tool designed to run and train large language models (LLMs) and diffusion models locally on Windows, macOS, and Linux. It offers a complete platform for research, model fine-tuning, export, and deployment from a single open-source application.

Key Features:

Runs and trains LLMs like Kimi K3, MiniMax-H3, Gemma 4, Qwen3.8, and DeepSeek-V4 Flash 0731.
Supports diffusion models with self-healing tool calling and sandboxed code execution for secure AI operations.
Includes private web search integration, enabling users to perform deep research and utilize retrieval-augmented generation (RAG).
Features modular component protocol (MCP) support and an OpenAI-compatible API for smooth integration with existing workflows.
Available via WinGet, making it easy to install and deploy using the winget command line.

Target Audience and Benefits: Ideal for researchers, developers, and businesses seeking to build and deploy AI solutions locally. Users benefit from increased control over their data while utilizing advanced AI capabilities in a user-friendly environment.