LM Studio
LM Studio is a desktop AI platform for discovering, downloading, running, and developing with open-source language models locally on Windows, macOS, and Linux, with private offline inference and local APIs.
LM Studio is a local AI platform designed for users who want to run large language models directly on their own computers instead of relying entirely on cloud-based AI services. It provides a graphical interface for discovering and downloading models such as Llama, Qwen, DeepSeek, Gemma, Mistral, and other compatible models, then running them locally through a familiar conversational interface.
Privacy is one of LM Studio's main advantages. Local inference keeps prompts, conversations, and model processing on the user's machine, and the free tier includes local LLM execution and offline voice transcription. Users can also attach documents and interact with them locally through retrieval-augmented generation (RAG), making LM Studio useful for private document analysis and knowledge workflows.
For developers, LM Studio can operate as a local AI server. It provides its own REST API alongside OpenAI-compatible and Anthropic-compatible endpoints, plus Python and TypeScript SDKs. This allows developers to replace or complement cloud model APIs with locally hosted models in applications, prototypes, agents, and development workflows. Its CLI can also manage models and servers, while headless operation makes it possible to run LM Studio without the desktop interface.
LM Studio also supports the Model Context Protocol (MCP), allowing local models to interact with external tools and services. Both local and remote MCP servers are supported, and newer integrations can use OAuth authentication. Developers can expose MCP functionality through LM Studio's APIs as part of agentic applications and tool-calling workflows.
More recently, LM Studio introduced Bionic, an AI agent designed for coding, research, document work, and other complex tasks using open models. Bionic can use models running locally, models available through another device via LM Link, or larger open models through LM Studio's cloud infrastructure.
LM Studio's core local experience is available for $0, including local LLMs and offline voice transcription. Cloud inference is available through pay-as-you-go credits, while additional Bionic subscription options are being introduced. Overall, LM Studio is especially useful for people who want greater privacy, control, and flexibility over how and where their AI models run.