What it is
LM Studio is a free desktop application for running language models locally. It has a built-in model browser (connected to Hugging Face), a chat window that looks like ChatGPT, and a one-click local server that other apps can use. It is the best starting point if you prefer not to use the terminal.
Install
- Download LM Studio for your system from lmstudio.ai.
- Windows: run the installer. macOS: open the
.dmgand drag LM Studio to Applications. Linux: download the AppImage, make it executable and run it. - Open the app. It checks your hardware and suggests models that fit.
Use it
- Open the Discover tab (magnifying glass) and search for a model, for example
gemma,qwenorllama. - Pick a version marked as fitting your machine (LM Studio shows a green or warning badge) and click Download.
- Go to the Chat tab, select the model at the top, and start typing.
- You can attach documents to a chat and ask questions about them.
Tip
On Macs with Apple Silicon, prefer MLX versions of models when available: they are usually faster.
Use the local server
Open the Developer tab and start the server. It listens on http://localhost:1234 with an OpenAI-compatible API, so apps and scripts written for OpenAI can point to it instead:
curl http://localhost:1234/v1/chat/completions -H "Content-Type: application/json" -d '{
"model": "your-model-name",
"messages": [{ "role": "user", "content": "Hello!" }]
}'LM Studio also comes with a command-line tool, lms, to load models and start the server from scripts.
Limitations
- The app itself is not open source (the models are open).
- Large models need a lot of memory; stick to the ones LM Studio says will fit.
Comments 0
Share tips, prompts that worked for you, or questions.
No comments yet. Be the first to share a tip.
Log in or create a free account to comment.