{"slug": "unsloth-desktop", "title": "Unsloth Desktop", "summary": "Unsloth released Unsloth Desktop (Beta), a free, open-source app for running and training AI models locally on macOS, Windows, and Linux, supporting LLMs, diffusion image/video, MLX, GGUF, and audio models. The app includes features like secure remote access via Cloudflare tunnel, integration with cloud models from OpenAI and Anthropic, and connection to agents like Claude Code and Codex, with no telemetry and offline capability.", "body_md": "**Introducing Unsloth Desktop**\n\n**Unsloth Desktop (Beta)** is a free, **open-source** app for running and training AI models on your own local hardware. Available for macOS, Windows, and Linux.\n\nUnsloth lets you run, train and deploy LLMs, **diffusion** image/video, **MLX**, GGUF and audio models.\n\n[Download Unsloth Desktop](https://unsloth.ai/download)[Features](/docs/desktop#features)[GitHub](https://github.com/unslothai/unsloth)\n\n**Features** ✨\n\n**Get started** 🦥\n\nInstall Unsloth Desktop, download a model, and start chatting in minutes.\n\n**Download Unsloth Desktop:**[macOS](https://unsloth.ai/download/mac)[Windows](https://unsloth.ai/download/windows)[Linux and WSL](https://unsloth.ai/download/linux)\n\n### Install Unsloth Desktop\n\nDownload\n\n[Unsloth Desktop](https://unsloth.ai/download)Launch the app\n\n### Unsloth is now ready\n\nTo chat, type a message and press Enter.\n\n**Connect tools:**[Claude Code](/docs/basics/claude-code),[Codex](/docs/basics/codex),[web search](/docs/new/studio/chat#advanced-web-search),[MCP](/docs/basics/mcp)and more**Train models:** Fine-tune text, diffusion,[embedding](/docs/basics/embedding-finetuning), and more\n\n**Feature Deep Dive ⭐**\n\nSee everything Unsloth Desktop has to offer:\n\n## Access model anywhere\n\nServe your local or Colab models over HTTPS through Unsloth's free [Cloudflare tunnel](/docs/basics/how-to-serve-local-llms-anywhere-secure-remote-access-with-cloudflare-and-unsloth). Check a run from **your phone**, your laptop, or anywhere else you happen to be.\n\nBind the app to your network with `-H 0.0.0.0`\n\n, or open a free Cloudflare tunnel for HTTPS:\n\n## Use Cloud Models\n\nRun models from OpenAI, Anthropic, Ollama, llama.cpp, vLLM, and more.\n\nUse the same Unsloth chat interface for local and cloud models with support for tool-calling, image gen, [prompt caching](https://sites.gitbook.com/preview/site_mXXTe/~/revisions/KsWgqmEjr66G4kYiZdG8#prompt-caching) to reduce token usage while preserving provider-native features like OpenAI’s [web search](https://sites.gitbook.com/preview/site_mXXTe/~/revisions/KsWgqmEjr66G4kYiZdG8#web-search-and-thinking) and [code execution](https://sites.gitbook.com/preview/site_mXXTe/~/revisions/KsWgqmEjr66G4kYiZdG8#code-execution).\n\n## Connect your Agent\n\n[Unsloth Start](/docs/integrations/unsloth-start) lets you connect [Claude Code](/docs/basics/claude-code), [Codex](/docs/basics/codex) and other agents to local models via the `unsloth start`\n\ncommand.\n\nStart Unsloth, load a model, open your project folder, and then run:\n\n#### Do you collect my data?\n\nNo telemetry. Unsloth detects your GPU type and device so the app can know what works. The app can run entirely offline.\n\n#### Can I use models I already downloaded?\n\nYes, they are found automatically. If yours are not you can specify your own custom folders.\n\n#### Why is inference slower sometimes?\n\nWeb search, code execution and tool-call healing all cost time. Turn them off and speed should match any other llama.cpp app. Still slow? Open a GitHub issue.\n\n#### GPU only?\n\nNo. Unsloth works on a wide variety of CPU, Mac etc setups.\n\n#### Does it support OpenAI-compatible APIs?\n\nYes. See the [API guide](/docs/basics/api). We also support connection to [Cloud models](/docs/integrations/connections) or other APIs like Anthropic or OpenAI.\n\n#### What devices does Unsloth support?\n\nUnsloth supports all OS including Mac, Windows, Linux and WSL and supports NVIDIA, Intel, AMD and Mac GPUs/CPUs. Older hardware however may not be well supported.\n\nA huge thank you to NVIDIA and Hugging Face for being part of our launch. Also thanks to all of our early beta testers for Unsloth Desktop, we truly appreciate your time and feedback. We’d also like to thank llama.cpp, PyTorch, stablediffusion.cpp, and open model labs for providing the infrastructure that made Unsloth Desktop possible.", "url": "https://wpnews.pro/news/unsloth-desktop", "canonical_source": "https://unsloth.ai/docs/desktop", "published_at": "2026-08-11 14:31:38+00:00", "updated_at": "2026-08-11 14:42:02.983143+00:00", "lang": "en", "topics": ["ai-tools", "ai-products"], "entities": ["Unsloth", "NVIDIA", "Hugging Face", "Claude Code", "Codex", "OpenAI", "Anthropic", "Cloudflare"], "alternates": {"html": "https://wpnews.pro/news/unsloth-desktop", "markdown": "https://wpnews.pro/news/unsloth-desktop.md", "text": "https://wpnews.pro/news/unsloth-desktop.txt", "jsonld": "https://wpnews.pro/news/unsloth-desktop.jsonld"}}