# LM Studio

> App for downloading and running AI models locally, with a Claude Code-compatible server

- Category: Agents & dev (`agents`)
- Official: https://lmstudio.ai/
- Wunderlust: https://wunderlust.xyz/r/lm-studio
- Image: https://wunderlust.xyz/media/lm-studio-poster.jpg
- Video preview: https://wunderlust.xyz/media/lm-studio.webm
- Added: 2026-11-05

## What it is

App for running language models locally; its server exposes a /v1/messages endpoint compatible with the Anthropic API.

## What it’s for

For running models on your own machine and pointing Claude Code or another agent at them.

## Benefits

- Local server with a /v1/messages endpoint compatible with the Anthropic API
- Downloads local LLMs in the app, with MLX and llama.cpp runtimes
- Link for using a remote model with encryption between devices

## Caveats

- Speed and context window depend on the GPU and configuration.
- Some endpoints are still in development.
- The official page doesn't state a price or license and highlights the Bionic agent, without mentioning the local server.

## Similar on Wunderlust

- [Ollama](https://wunderlust.xyz/r/ollama.md): Runtime for running language models on your own machine
- [Unsloth Studio](https://wunderlust.xyz/r/unsloth-studio.md): App for running, chatting with and training AI models locally
- [Qwen 3.5](https://wunderlust.xyz/r/qwen-3-5.md): Qwen 3.5 model family for running on your own hardware
- [OmniRoute](https://wunderlust.xyz/r/omniroute.md): An open source gateway that routes AI tools through an OpenAI-compatible endpoint
- [MLX](https://wunderlust.xyz/r/mlx.md): Stack for running local models on Macs
- [Claude Fable 5.1](https://wunderlust.xyz/r/claude-fable-5-1.md): Anthropic model for code, knowledge work and long-running tasks