Product overview

What is LM Studio?

LM Studio is software for discovering and running AI models on a user's own computer or server. The desktop application supports local models such as gpt-oss, Qwen, Gemma, DeepSeek, and others, allowing prompts and model inference to run on local hardware. It is free for home and work use under the product's terms.

For server and automated environments, LM Studio provides llmster, a headless version of its runtime that can be deployed on Linux machines, cloud servers, or CI without a graphical interface. Developers can integrate local models through JavaScript and Python SDKs, an OpenAI-compatible API, the lms command-line interface, and MCP client support.

How to Use LM Studio

Download and install the desktop application for the supported operating system.
Explore available models and select one that fits the machine's resources and task.
Download the model and load it in LM Studio.
Chat with the model locally or expose it through the local API for another application.
For servers, install the headless runtime and manage models without the desktop interface.

Core Features

Local model runtime: Runs supported language models on the user's own hardware.
Desktop interface: Provides model discovery, download, loading, and interaction in one app.
Headless deployment: Uses llmster on Linux, cloud servers, or CI without a GUI.
Developer SDKs: Offers JavaScript and Python packages for application integration.
OpenAI-compatible API: Lets compatible clients call models served by LM Studio.
CLI and MCP support: Includes the lms command line and an MCP client.

Use Cases

Private local chat: Individuals can run prompts without sending model inference to a hosted provider.
Local application development: Developers can test model-powered features through SDKs or the compatible API.
Offline experimentation: Teams can explore downloaded models on local hardware.
Server inference: Organizations can run the headless runtime on Linux or cloud infrastructure.

Pricing

The website states that LM Studio is free for home and work use, subject to its terms.

Frequently Asked Questions

Does LM Studio require a cloud model provider?

No. Its primary purpose is to run downloaded models locally on the user's hardware.

Can LM Studio run without a desktop interface?

Yes. The llmster runtime supports headless deployment on Linux, cloud servers, and CI.

Back to product directory

LM Studio