LM Studio
LM Studio is a desktop application for downloading, running, and serving local large language models (LLMs) on macOS, Windows, and Linux.
LM Studio provides a built-in model catalog and importer for popular local formats such as GGUF and MLX, along with an offline-first chat interface. The app also supports attaching files to chats to add context through retrieval-augmented generation, which makes it useful for local document Q&A without sending data to external services.
For developers, LM Studio can expose locally loaded models over OpenAI-compatible APIs for chat completions and embeddings, as well as an Anthropic-compatible /v1/messages endpoint, and it offers official Python and TypeScript SDKs. LM Studio also supports the Model Context Protocol, or MCP, letting you connect MCP servers to your local models so they can use external tools. It also offers an lms command-line tool for scripting tasks like loading models and starting the local server. LM Link extends the same local server across devices, letting you reach the models loaded on one machine from another over an end-to-end encrypted connection. The CLI and SDKs include controls for load-time configuration such as context length and GPU offload. LM Studio can run headless as a background service, so you can serve models on machines with no desktop session.
The same team also ships LM Studio Bionic, a separate agent app built on the LM Studio runtime for coding, research, and document work with open models.
Official website: lmstudio.ai
Related Resources
Tutorial
How to Use Ollama to Run Large Language Models Locally
Learn how to use Ollama to run large language models locally. Install it, pull models, and start chatting from your terminal without needing API keys.
For additional information on related topics, take a look at the following resources:
