LM Studio
Run AI models on your own computer, free and offline
Free desktop app for running open AI models like Llama, Gemma and Qwen locally, with no account and no data leaving your computer.
Independent listing: we have no partnership with this tool.Independent listing, no partnership.Independent listing. Learn more
About LM Studio
LM Studio lets you run AI chat models on your own computer. You pick a model, like Llama, Gemma, Qwen or Mistral, download it inside the app and chat with it the same way you would with ChatGPT. Once the model is downloaded, everything works offline.
Because nothing goes to the cloud, it is a good choice for private documents, client work or just trying open models without paying per message. Developers can also start a local server that speaks the same language as the OpenAI API and plug it into their own apps.
The app is free for personal use and for work. How well it runs depends on your computer: more memory and a good graphics card mean bigger and faster models.
Best for
- Developers
- Privacy-minded users
- Students
- Researchers
- Small businesses with sensitive data
Key features
- Download and run open models in one app
- Works fully offline after download
- Chat with your own documents
- Local server compatible with the OpenAI API
- Runs on Windows, macOS and Linux
- Apple Silicon support with MLX
Pros & cons
Pros
- Completely free, also for work
- Your data never leaves your computer
- No account or subscription needed
- Easy way to try many open models
Cons
- Needs a fairly strong computer for big models
- Local models are weaker than the best cloud models
- Large downloads, often several GB per model
Pricing plans
Full desktop app for personal and work use, Windows, macOS and Linux
SSO, model controls, private sharing and priority support
Prices are shown in USD and can change. They may also vary by country, currency, taxes and billing period (monthly or yearly). Always confirm the final price on the official website.
