Desktop app for running local LLMs with a chat UI and API
LM Studio is a desktop application for downloading and running open-weight language models on your own machine. It provides a model browser backed by Hugging Face, a chat interface, and a local server that mimics the OpenAI API so existing code can point at localhost. The pitch is privacy and offline use: prompts and documents never leave the computer. It runs on Windows, macOS and Linux, uses whatever GPU or Apple Silicon hardware is present, and works without a cloud account, which is the practical difference from hosted chat tools.
Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.
private chat, offline llms, local prototyping, embedded model testing
If you're comparing similar products, check the alternatives below, or browse all tools in the AI Models & Platforms category.
No. Models run locally on your machine, and conversations stay on disk. The app does contact Hugging Face when you search for and download models, but your prompts and documents are not uploaded.
A modern GPU with 8GB or more VRAM, or an Apple Silicon Mac with 16GB or more unified memory, handles small models well. Larger models need proportionally more memory and run slowly on CPU only.
The desktop application is free for personal use. Commercial use of the app itself requires a paid licence, though the models you download carry their own separate licences you should also check.
Developer console and API keys for the Claude model family
Node-based local interface for running image and video diffusion models
Browser playground for prompting and prototyping with Gemini models
Hosted Jupyter notebooks with optional free GPU and TPU runtimes