

LM Studio
Enables running large language models entirely offline on desktop with integrated chat UI, OpenAI-compatible server, HuggingFace downloads, model discovery, support for multiple LLM formats and architectures, local privacy, GPU/CPU operation, and quick setup.
Cost / License
- Free
- Proprietary
Application types
Platforms
- Mac
- Windows
- Linux
Features
- Ad-free
- Works Offline
- No registration required
- No Tracking
- Dark Mode
- AI Chatbot
- No Coding Required
LM Studio News & Activities
Recent News
- Fla published news article about LM Studio
LM Link launches on iPhone, bringing local AI model access to iOS devicesLM Link is now available on iPhone and iPad, allowing users to run large local AI models directly f...
- POX published news article about LM Studio
LM Studio 0.4 adds parallel model requests, server-native daemon and new stateful REST APILM Studio 0.4 has been released, expanding its platform for discovering and running local open larg...
- Maoholguin published news article about LM Studio
LM Studio is now free for both personal and commercial use at work!LM Studio is now free to use in professional settings, removing the need for separate licenses or p...
Recent activities
Featured in Lists
Local AI is a curated list of software that lets you run and use AI directly on your own computer — without relying on …
What is LM Studio?
With LM Studio, you can ... 🤖 - Run LLMs on your laptop, entirely offline 👾 - Use models through the in-app Chat UI or an OpenAI compatible local server 📂 - Download any compatible model files from HuggingFace 🤗 repositories 🔭 - Discover new & noteworthy LLMs in the app's home page








Comments and Reviews
Quite literally the only GPU & CPU inferencing app self host that isn't a lie or complete paywall bs. Intergrates with Huggingface to find your preferred ggufs, Why can't other people be like this. Thanks LM Studio for the only no BS app to run LLMs.
It's very suitable for beginners, easy to use, and has a clean and intuitive interface. Updating the program itself and its dependencies (like llama.cpp, CUDA, etc.) is also quite straightforward. It supports GPU acceleration, and you can easily configure the model during loading. However, it lacks a 'Chat with Documents' feature, and therefore does not support capabilities like RAG and embedding. Plus not open source.
I would give it a 5 if it wasn't for an update that removed the option to select the chats folder, but for everything else have no complaint whatsoever, but just for that detail, I'm thinking about switching to another software alike
Easy to use, know if the model could be run on our computer at a glance. Also, easy to deploy a local server. What else ?