How to Install LM Studio on Windows, Mac, and Linux (A Beginner's Guide)

Published on August 12, 2026 • By LMSA
How to Install LM Studio on Windows, Mac, and Linux (A Beginner's Guide)

If you've ever wanted to chat with an AI model without sending your data off to some company's server, LM Studio is one of the easiest ways to get there. It's a free desktop app that lets you download open-source language models and run them entirely on your own computer — no internet connection required once a model is downloaded, no subscription, and no one looking over your shoulder.

The good news is that installing it is genuinely simple, even if you've never touched a terminal in your life. This guide walks you through the process on Windows, Mac, and Linux, step by step. By the end, you'll also know how to reach your models from your phone using a free companion app called LMSA — handy for chatting on the couch without hauling your laptop around.

What You'll Need Before You Start

LM Studio itself is a lightweight download (usually somewhere between 150 and 450 MB depending on your platform), but the AI models you'll run inside it can be several gigabytes each. Before you install anything, make sure you have:

  • At least 10 GB of free disk space (more if you plan to try multiple models)
  • 16 GB of RAM as a comfortable starting point, though smaller models can run on less
  • A relatively modern CPU or GPU. On Windows, your processor should support the AVX2 instruction set, which almost anything made in the last several years already does
  • macOS 13.4 or later if you're on a Mac, and Ubuntu 20.04 or later (or an equivalent modern Linux distribution) if you're on Linux

Don't worry too much about the specifics — LM Studio will tell you what your hardware can handle once it's installed, and it's smart about recommending models that fit your setup.

Step 1: Download LM Studio From the Official Site

Regardless of which operating system you use, everything starts in the same place: lmstudio.ai. Open your browser and head there directly rather than searching for a third-party mirror — it only takes one fake download link to end up with something you didn't want on your computer.

The homepage will detect your operating system automatically and show you the right download button, but if it doesn't, you'll find separate options for Windows, macOS (Apple Silicon and Intel), and Linux.

A quick note for Mac users: if you have an M-series chip (M1 through M5), make sure you grab the Apple Silicon build rather than the Intel one. The Apple Silicon version taps into Metal for GPU acceleration, which makes a real difference in speed. Running the Intel build through Rosetta will still work, but it's noticeably slower.

Step 2: Install LM Studio on Windows

  1. Once the .exe file finishes downloading, find it in your Downloads folder and double-click it.
  2. Windows may show a security prompt (SmartScreen) since the app isn't signed by a major publisher you already trust. Click More info, then Run anyway to continue.
  3. The installer wizard will ask where you'd like to install the app. The default location is fine for most people — just make sure whichever drive you pick has enough free space.
  4. Click through the prompts to finish. You typically don't need administrator rights for this.
  5. Once it's done, launch LM Studio from the Start menu or the desktop shortcut it created.

That's it — you're in.

Step 3: Install LM Studio on Mac

  1. Open the downloaded .dmg file from your Downloads folder.
  2. A window will pop up showing the LM Studio icon and a shortcut to your Applications folder. Drag the icon into Applications.
  3. Open your Applications folder and double-click LM Studio to launch it.
  4. macOS may warn you that the app was downloaded from the internet and ask you to confirm you want to open it. Click Open.
  5. If macOS blocks it entirely (this can happen with Gatekeeper on stricter security settings), go to System Settings → Privacy & Security, scroll down, and click Open Anyway next to the LM Studio entry.

From here, LM Studio should launch normally every time you click its icon.

Step 4: Install LM Studio on Linux

LM Studio for Linux ships as an AppImage, which is a self-contained file that doesn't need a traditional installation process. Here's how to get it running:

  1. Move the downloaded AppImage file to a folder where you'd like to keep it (many people use ~/Applications or simply their home folder).

If it doesn't launch, you may be missing FUSE, which many minimal Linux installs don't include by default. On Ubuntu or Debian-based systems, you can install it with:

sudo apt install libfuse2

Double-click the file, or run it from the terminal:

./LM-Studio-*.AppImage

Make it executable. You can right-click the file, choose Properties, and enable "Allow executing file as program," or do it from a terminal:

chmod +x LM-Studio-*.AppImage

Once it opens, you're ready to go, and you can create a desktop shortcut manually if you'd like it to show up in your applications menu.

Step 5: Download Your First Model

With LM Studio open on any platform, the next step is the same everywhere. Click the search or "Discover" icon (usually a magnifying glass in the sidebar) to browse available models. LM Studio pulls these from Hugging Face, so you'll see well-known names like Llama, Mistral, Phi, Gemma, and Qwen.

If you're not sure where to start, look for something in the 3B to 8B parameter range with a "Q4" quantization label — these are smaller, faster models that run comfortably on most laptops. A model like Phi-4 Mini or Llama 3.2 3B is a solid, lightweight first download.

Click download, wait for it to finish, then load it from the chat tab. Within a few minutes of installing LM Studio, you can be having a full conversation with an AI model that never leaves your machine.

Bonus Step: Chat With LM Studio From Your Phone Using LMSA

Here's something a lot of beginners don't realize right away: LM Studio can run a local server, which means other devices on your Wi-Fi network can talk to it too — including your Android phone.

This is where LMSA comes in. LMSA is a free Android app built specifically to be a mobile front-end for LM Studio. Instead of being stuck at your desk to chat with your models, you can open LMSA on your phone and talk to the same model running on your computer, from anywhere in the house.

Setting it up takes just a few minutes:

  1. In LM Studio, go to the Developer tab (the </> icon) and turn on Serve on Local Network, along with the local server toggle. Make sure a model is loaded.
  2. Find your computer's local IP address (on Windows, run ipconfig in Command Prompt; on Mac or Linux, use ifconfig or check your network settings).
  3. Install LMSA on your Android device from the Play Store.
  4. Open LMSA, go to its settings, and enter your computer's IP address and port (1234 by default).
  5. Connect your phone to the same Wi-Fi network as your computer, and start chatting.

LMSA is built with privacy in mind — it doesn't track you, and everything stays on your local network rather than passing through a third party. It also supports connecting to Ollama and, if you want cloud models too, OpenRouter, so it's a nice all-in-one mobile chat client if your local AI setup grows over time.

Some Final Notes

Installing LM Studio takes only a few minutes on any platform, and it opens the door to running capable AI models entirely offline, on hardware you already own. Once you're comfortable with the desktop app, pairing it with LMSA turns your phone into a remote control for your own private AI — no cloud subscription, no data leaving your home network, and no one else's server standing between you and your conversations.

Take it one step at a time: install the app, download a small model, get a feel for how it runs, and then expand from there. Local AI has gotten remarkably approachable, and you now know exactly how to get started.