toolsy

Categories

AI & Automation
Translation
Gaming
Media Tools
Customization
System Utilities
Security

Language

LM Studio

Run AI Models Locally With GUI — LM Studio

Run AI models like Llama, Mistral, and Gemma locally with a GUI — no terminal, no cloud, no subscription required.

v0.3.6
May 2026
WindowsmacOSLinuxFreeOpen Source

LM Studio is a desktop application that lets you download and run large language models directly on your computer — without writing a single command. Where tools like Ollama require a terminal, LM Studio gives you a clean graphical interface to browse, download, and chat with hundreds of AI models.

Install LM Studio, open it, and you can search and download models like Llama 3, Mistral, Phi-3, Gemma, DeepSeek, Qwen, and many others directly from within the app. No manual downloads, no configuration files — just click and run. Models are stored locally and work completely offline once downloaded.

LM Studio also includes a built-in chat interface that mimics the ChatGPT experience, with support for conversation history, system prompts, and model switching. You can set parameters like temperature, context length, and GPU layers through sliders — no JSON editing required.

For developers, LM Studio exposes a local OpenAI-compatible REST API server. Any application that works with the OpenAI API can be pointed at LM Studio instead, enabling local AI integration without code changes. This makes it easy to use with tools like Cursor, Continue, or custom scripts.

Performance is strong on Apple Silicon Macs thanks to Metal GPU acceleration. On Windows and Linux, NVIDIA GPUs with CUDA are supported, with CPU-only fallback for machines without a compatible GPU. Smaller models like Phi-3 Mini or Gemma 2B run well even on modest hardware.

LM Studio is free to use for personal and research purposes. No account required, no data sent to any server during inference.

Related Tools

  • Ollama — Terminal-based local AI runtime with REST API, ideal for developers and integrations
  • Flowise — Build no-code AI workflows powered by local models running in LM Studio

FAQ

Is LM Studio free?+
Yes, LM Studio is free to use for personal and research purposes. It is not open source, but there is no cost, no subscription, and no account required. You can download and run hundreds of AI models at no charge.
LM Studio vs Ollama — which should I use?+
LM Studio is better if you want a graphical interface — click to download models, chat in a built-in UI, adjust settings with sliders. Ollama is better if you are a developer who wants a lightweight background service with a REST API to connect to other tools. Both run models locally and work offline.
What AI models can I run in LM Studio?+
LM Studio supports any GGUF-format model, which includes Llama 3, Mistral, Phi-3, Gemma, DeepSeek, Qwen, Code Llama, and hundreds of others. You can browse and download them directly from within the app via Hugging Face integration.
Does LM Studio work without internet?+
Yes. Once you have downloaded a model, LM Studio runs it entirely offline. No data is sent to any server during inference. Internet is only needed for the initial model download.