LM Studio

A desktop app to discover, download, and run local LLMs.

Hand-reviewed by the Toolglade team·Updated Aug 2026

coding#local-llm#desktop-app#privacy#open-weight

Free plan Claimed API Self-hosted

Toolglade’s take

LM Studio is the friendliest on-ramp to local LLMs for people who want a GUI: browse models, download, chat, and expose a local OpenAI-compatible server, all without touching a terminal. It is free for personal and commercial use, which removes a common friction point. It suits developers evaluating models, privacy-focused users, and teams building on-device features. Caveats are inherent to local inference: your hardware sets the ceiling on speed and model size, and organizations wanting SSO and central controls need the paid Enterprise tier. A great desktop companion to command-line tools like Ollama.

About LM Studio

LM Studio is a desktop application that makes running open-weight LLMs locally approachable for people who prefer a graphical interface over the command line. It lets you browse and download models (from sources like Hugging Face), chat with them in a built-in UI, manage quantization and hardware settings, and spin up a local OpenAI-compatible server so your own apps can call models running on your machine. It runs on macOS, Windows, and Linux and supports popular local inference backends, handling model loading, GPU offload, and context settings through a polished interface. This makes it a common choice for developers testing models, privacy-conscious users, and anyone who wants offline AI without cloud dependencies or per-token costs. Since mid-2025 LM Studio has been free for both personal and commercial use with no separate license required, and the company offers a separate paid Enterprise tier for organizations needing features like SSO, model gating, and centralized controls. The main limits are the usual local ones: performance depends on your hardware, and the largest models require substantial RAM and a capable GPU.

TL;DR

LM Studio is a free desktop app for discovering, downloading, and running open-weight LLMs locally through a friendly GUI, with a built-in chat interface and a local OpenAI-compatible server. It is free for personal and commercial use and targets people who prefer a graphical experience over the command line. It suits developers evaluating models and privacy-conscious users. Its limits are those of local inference: hardware sets the ceiling, and organizations needing central controls require a paid Enterprise tier. It pairs well with CLI tools like Ollama.

Company overview

LM Studio is developed by Element Labs (Element Labs, Inc.), the company behind the app, founded by its original creator. It focuses on making local LLMs accessible through a polished desktop experience.

In mid-2025 the company made the app free for personal and commercial use, removing the prior workplace license requirement, and introduced a separate Enterprise offering for organizations.

Product features

LM Studio provides a GUI to browse, download, and run open-weight models locally, with a built-in chat interface and controls for quantization, GPU offload, and context. It runs a local OpenAI-compatible server for application integration.

It supports macOS, Windows, and Linux and popular local inference backends. A separate Enterprise tier adds SSO, model and MCP gating, and centralized administration for organizations.

Target market

Developers, privacy-conscious users, and newcomers to local AI who want a graphical way to run open-weight LLMs offline, plus organizations that need centralized controls via the Enterprise tier.

Buyer personas

End users

Developers and enthusiasts running open-weight models locally through a GUI.

Buyers

Individuals (free) and organizations purchasing the Enterprise tier.

Key influencers

Local-AI community members and privacy advocates.

Ideal customer profile

A developer or privacy-focused user who wants a friendly desktop app to run local LLMs offline, with an optional enterprise path for organizational controls.

Funding & performance

Funding reports for LM Studio are mixed: some sources describe it as bootstrapped, while others report roughly $19 million raised from investors including Conviction and others. Because reports conflict, treat funding figures as unconfirmed and verify with the company.

Pros & cons

Pros

Polished, beginner-friendly GUI
Free for personal and commercial use
Local OpenAI-compatible server for integration
Cross-platform support
Handles quantization and GPU offload settings
Easy model discovery and download
Runs fully offline for privacy

Cons

Performance bounded by local hardware
Largest models need lots of RAM and a strong GPU
No built-in team collaboration in the free app
Enterprise features require a paid tier
Not designed for server-scale multi-user serving
Model quality depends on user's choices

Pricing plans

Free

Free for personal and commercial use
Download and run open-weight models
Built-in chat GUI
Local OpenAI-compatible server

Enterprise

Custom

SSO
Model and MCP gating
Centralized administration
Priority support

Key features

API

Self-hosted

Multi-language

Integrations

OpenAI-compatible local server, Hugging Face models, LangChain, llama.cpp, MLX

Input types

text, image

Output types

text

Best For

Running local LLMs with a GUI, Testing and comparing models, Offline private chat, Local API for app development

Compare key features

View all alternatives →

Feature

LM Studio

Ollama

SuperCompress

Pricing

Freemium

Free plan

Yes

Free trial

API

Yes

Self-hosted

Yes

Frequently asked questions

Is LM Studio free?+

Yes. The desktop app is free for both personal and commercial use, with no separate license required. A paid Enterprise tier exists for organizations.

Do I need to use the command line?+

No. LM Studio provides a graphical interface for downloading, running, and chatting with models, which is its main appeal versus CLI tools.

Can my apps use models running in LM Studio?+

Yes. It can run a local OpenAI-compatible server so your applications can send requests to models running on your machine.

What operating systems are supported?+

macOS, Windows, and Linux, with support for backends like llama.cpp and MLX depending on your platform.

What hardware do I need?+

Smaller quantized models run on modest machines, but larger models require substantial RAM and a capable GPU for good performance.