LM Studio
Desktop app for discovering, downloading and running open models on a personal machine, with a local OpenAI-compatible server other apps can call.
Find alternatives to LM Studio
What it does
- Model inference
- Model hosting
Features 17
- Chat interfaceA simple and flexible chat interface for running local LLMs.
- MCP clientInstall MCP servers in LM Studio and use them with local models.
- Model search and downloadSearch and download models from within the app via Hugging Face.
- Local OpenAI-like serverServes local models on OpenAI-like endpoints, locally and on the network.
- Chat with documents offlineAttach documents to chat messages and interact with them entirely offline (RAG).
- lms CLICommand-line tool for chat, model downloads, daemon management, server control and publishing.
- Python and TypeScript SDKslmstudio-python and lmstudio-js SDKs for loading models, generating text, embeddings and plugins.
- PresetsManage local models, prompts and configurations as presets that can be imported, pushed and pulled.
- Split view in chatViews two chats side by side by dragging chat tabs to either half of the window.
- Speculative decodingIncreases generation speed without reducing response quality.
- Parallel requests via continuous batchingCombines multiple requests into a single batch on the server.
- Just-in-time model loadingLoads models on demand so other apps can use them.
- Idle TTL and auto-evictSets how long a model stays loaded without requests.
- Add to LM Studio deeplink buttonInstalls MCP servers in one click through a deeplink.
- API token authenticationSecures access to the LM Studio API with API tokens.
- Chat exportExports chats to PDF, markdown or plain text.
- Developer ModeCombines the Developer and Power User modes with all advanced features enabled.
How it compares
-
AnythingLLM
AnythingLLM vs LM Studio: the open-source LM Studio alternative
Company disclosure · 13 Oct 2026
Sources
- Pricing
-
'Free — $0 — Run local LLMs on your machine', quoted from the page, 2026-09-18.
- Description
-
Official documentation · 18 Sep 2026
The documentation's own framing: 'Learn how to run Llama, DeepSeek, Qwen, Phi, and other LLMs locally with LM Studio.'
- Constraints
-
Official documentation · 2 Oct 2026
Derived from the security facts on this page, from the same source.
- Security
-
Official documentation · 2 Oct 2026
Privacy policy plus enterprise page. Training/ZDR statements concern cloud features (cloud models, web search) and the ZDR cloud; the local app sends no chat content. No certifications, residency options or admin features stated; no trust centre.
- Sold within
-
Downloaded and run on its own; the paid tiers are hosted usage, not a licence for the app.
- URL
-
Official documentation · 18 Sep 2026
Fetched 2026-09-18, 200.
- Interfaces
-
Official documentation · 14 Oct 2026
Page: 'LM Studio is available for macOS, Windows, and Linux'.
- Modalities
-
Official documentation · 1 Oct 2026
Page: 'Chat and text generation with streaming / Create embeddings, inspect tokens / Generate typed JSON outputs that validate against a schema'.
Also from LM Studio
-
LM Studio Bionic AI agent
LM Studio's agent for open models, run locally for creative, knowledge and coding work, with optional US-hosted open models on the paid tiers.
-
LM Link Infrastructure service
LM Link lets users access local models across their devices.
-
llmster Infrastructure service
llmster is a standalone daemon that runs without a graphical interface.
Open LM Studio in the directory
Something wrong here? Send a correction — quote this product id: lm-studio.