LM Studio

LM Studio is a desktop app for Windows, Mac and Linux that lets you search for, download and use AI models on your computer from a window, without typing commands in the terminal. It's probably the friendliest way in for trying local AI for the first time. The difference with Ollama is one of approach: LM Studio is designed to be used with the mouse, and Ollama for the terminal and for other programs to use behind the scenes.

What it offers is pretty intuitive: a model browser that tells you whether a model will fit on your machine, a chat similar to ChatGPT's, the ability to work with your documents and a server mode so other apps can connect to it. What usually confuses beginners is choosing the model: the same one appears in several sizes and several quantized versions, that is, compressed to take up less space in exchange for some quality. It's best to start with a small one and work your way up. And beware: a comfortable window doesn't make the model any smarter; that depends on the model and your hardware.

LM Studio is one of the four harnesses I use locally, along with Ollama, llama.cpp and Unsloth; I call a harness whatever makes a model usable. In "Ollama, LM Studio, llama.cpp and Unsloth: which one I use for what" I explain what I use each one for, the real problems I've run into with them and which one I'd recommend to a beginner, to someone after performance and to someone who wants to fine-tune a model.