Unsloth Desktop

Unsloth is presented as an open-source, no-code desktop UI for training and running open models on local hardware. The page describes Unsloth Desktop as a local AI workspace for running, training, and connecting text, image, video, and audio models. It states the app is free, open-source, and 100% local, with downloads for Mac (Apple Silicon and Intel), Windows 10 and later, and Ubuntu and compatible Linux distros. Listed capabilities include image and video generation, connecting agents such as Claude Code and Codex via the unsloth start command, an OpenAI-compatible API, built-in private web search with Deep Research, sandboxed Bash and Python execution for models, a built-in model hub for choosing quantizations, and serving local or Colab models over HTTPS through a free Cloudflare tunnel.

Key features

• Image and video generation with listed models such as Qwen-Image-2.1, FLUX, Wan, and LTX
• Image transformation, inpainting, extension, upscaling, and editing locally
• Connect Claude Code, Codex and other agents to local models via the unsloth start command
• OpenAI-compatible API for existing apps, scripts, and SDKs
• Built-in web search and Deep Research with citations
• Self-healing tool calls and a secure sandbox for executing Bash and Python
• Built-in model hub to discover, manage, and download quantizations
• Serve local or Colab models over HTTPS through a free Cloudflare tunnel
• No-code workflows for training and generating media

Best suited for

• Running and chatting with models locally
• Training models with no-code workflows
• Generating images and video on local hardware
• Connecting coding agents to local models
• Private local web research with citations

Pricing

Free and open-source, as stated on the website.