LM Studio

LM Studio

Discover, download, and run local LLMs — a powerful desktop app for on-device AI.

Desktop app to run open LLMs locally on macOS, Windows, and Linux — chat UI, OpenAI-compatible local server, offline transcription. Free plan.

LM Studio homepage — local LLM desktop app

LM Studio homepage — local LLM desktop app

Overview

What is LM Studio?

Discover, download, and run local LLMs — a powerful desktop app for on-device AI.

LM Studio is a desktop application for running open large language models locally on your own computer. Its Bionic app is a local AI agent for creative work, coding, and documents that runs downloaded models on-device via the LM Studio runtime (llama.cpp and MLX), with optional pay-as-you-go cloud inference for heavier tasks.

The Free plan ($0) covers the Bionic Agent, local LLMs, offline voice transcription, and LM Link for up to 5 devices. Bionic+ ($20/month) and Pro ($100/month) add cloud inference usage; LM Studio itself is freeware, not open-source.

Platforms and languages

LM Studio Availability

Platforms

WindowsMacosLinux

Languages

English

Capabilities

LM Studio Key Features

Local model library

Download and run GGUF (llama.cpp) and MLX models directly inside the app.

Chat UI

Point-and-click chat interface — no terminal required.

Local OpenAI-compatible server

Programmable server at http://localhost:1234/v1 for apps and coding tools.

Cross-platform

macOS Apple Silicon, Windows (x64/ARM), and Linux (AppImage).

Offline voice transcription

Fully local, real-time voice transcription — voice data never leaves the device.

Best for

Who uses LM Studio?

Individuals and developers

Chat with open LLMs privately on-device and connect apps via a local API

Plans and access

LM Studio Pricing

Freemium

Free $0 (Bionic Agent, local LLMs, offline transcription). Bionic+ $20/mo, Pro $100/mo. Cloud usage billed via credits.

No free trial listed

Common questions

LM Studio FAQs

Is LM Studio free?

Yes — the Free plan ($0) includes the Bionic Agent and local model support. Bionic+ ($20/mo) and Pro ($100/mo) add cloud inference.

Can I run models fully locally without the cloud?

Yes. Local inference runs on your machine; credits are only consumed for cloud models.

Which platforms are supported?

macOS on Apple Silicon (M1–M4), Windows (x64 and ARM), and Linux (x64/ARM64 via AppImage). 16GB+ RAM recommended.

How do apps connect to LM Studio?

Via its OpenAI-compatible local server at http://localhost:1234/v1.

Reviews

See what the community thinks and share your experience.

—

Based on 0 ratings

Rating distribution

0
0
0
0
0

Leave a review

Sign in to rate

Community reviews

No reviews

No written reviews yet.

Explore by category

AI Desktop Apps

View all AI Desktop Apps websites
Burnstate preview
Burnstate

macOS menu-bar app that shows every AI coding limit in one place and warns you before you run dry.

AI Desktop Apps
Binqo preview
Binqo

A tiny AI friend called Mielo who lives in your Mac's notch — music, files, focus, weather and answers out loud, right where you already look.

AI Desktop Apps
RigSpark preview
RigSpark

Checks your RAM, GPU and disk, then picks the best AI model that actually runs well on your computer — free, open source and fully offline.

AI Desktop Apps
Voyager preview
Voyager

Voyager is a macOS file browser that brings AI agents into everyday file work, letting you find files and use them as context for Claude Code and Codex.

AI Desktop Apps
Kimi Work preview
Kimi Work

Moonshot's desktop AI agent for knowledge workers: local files, browser tasks, and scheduled jobs, powered by Kimi K3.1 Pro.

AI Desktop Apps

Explore by category

Llms Foundation Models

View all Llms Foundation Models websites
Sarvam Prism preview
Sarvam Prism

Sarvam AI's open-weight System One decision model that scores fixed options in a single pass and returns calibrated probabilities for enterprise workflows.

Llms Foundation Models

Explore by category

AI Development Infrastructure

View all AI Development Infrastructure websites
Plurai preview
Plurai

Simulation and evaluation platform that tests AI agents in sandboxed digital twins, finds failures before users do, and closes the fix loop.

AI Development Infrastructure