Install and Chat with LM Studio
Ollama is great, but if you would rather never see a terminal at all, LM Studio is for you. It is a free desktop app with a full graphical interface: you browse a catalog of models, click to download, click to load, and start chatting in a clean window. It is the most beginner-friendly way to run local AI, and many people prefer it as their main tool.
In this lesson you will install LM Studio, download your first model through its built-in catalog, and have a private conversation, all with your mouse.
What You'll Learn
- How to install LM Studio on Windows or Mac
- How to find and download a model from its built-in catalog
- How to load a model and chat with it
- How LM Studio compares to Ollama so you can pick a favorite
Step 1: Download and Install
Go to lmstudio.ai and download the version for your operating system (Windows, Mac, or Linux). Open the downloaded file and install it like any other app. LM Studio is free for personal use, and the makers have stated it is also free for commercial use, so you do not need to worry about a license for work tasks.
When you open LM Studio for the first time, you get a clean home screen. There is no background service to configure and nothing to set up in a terminal. Everything happens inside this one window.
Step 2: Find and Download a Model
LM Studio has a built-in model browser connected to Hugging Face, the large public home of open models. This means you can search and download models without leaving the app.
- Click the search or discover icon (a magnifying glass, usually in the left sidebar).
- Search for a model family. Good beginner choices to search are "Llama 3.1 8B," "Gemma," "Mistral," or "Phi."
- LM Studio shows the available versions. It often marks which ones are likely to run well on your machine, and it shows file sizes so you can match your memory from the earlier lesson.
- Click Download on a version that fits. A small or mid-size model (around 7-8B) is a solid first pick for most laptops.
The download runs inside the app with a progress bar. This is the one-time step; afterward the model lives on your disk.
Step 3: Load the Model and Chat
- Go to the chat icon (usually a speech bubble in the sidebar).
- At the top, click to select a model to load and choose the one you just downloaded. Loading places the model into your memory so it is ready to answer.
- Type a message in the box at the bottom and press Enter.
The reply is generated on your own hardware. Try "summarize the plot of Romeo and Juliet in three sentences" or "help me write a polite reminder email about an overdue invoice." When you are done, you can unload the model to free up memory, or just close the app.
- Search catalogBuilt-in model browser
- DownloadOne-time, inside the app
- Load modelInto memory, one click
- ChatPrivate, on your machine
Ollama or LM Studio: Which Should You Use?
You do not have to choose only one; many people keep both installed. But here is how they compare so you know when to reach for each.
Both are free and private. Ollama is lean and command-friendly; LM Studio is the most point-and-click.
| Criteria | Ollama | LM Studio |
|---|---|---|
| Main way to use it | One terminal command, plus a simple chat window | Full graphical app, all mouse |
| Finding models | A web library you browse | A search catalog built into the app |
| Best for | Quick starts, people okay with a command | Total beginners who want zero terminal |
| Cost | Free, open source | Free, incl. commercial use |
| Everyday chat | Yes | Yes |
| Runs offline after download | Yes | Yes |
Ollama
- Main way to use it
- One terminal command, plus a simple chat window
- Finding models
- A web library you browse
- Best for
- Quick starts, people okay with a command
- Cost
- Free, open source
- Everyday chat
- Yes
- Runs offline after download
- Yes
LM Studio
- Main way to use it
- Full graphical app, all mouse
- Finding models
- A search catalog built into the app
- Best for
- Total beginners who want zero terminal
- Cost
- Free, incl. commercial use
- Everyday chat
- Yes
- Runs offline after download
- Yes
A simple rule: if you want the most guided, mouse-only experience, start with LM Studio. If you like a lean tool and do not mind typing one command, Ollama is excellent. Trying both takes ten minutes and helps you find your preference.
A Note on Model Names
Inside LM Studio you may see the same model offered in several versions with codes like "Q4" or "Q8" in the name. Those refer to compression levels, and picking the right one is the subject of the very next lesson. For now, if LM Studio suggests a version as a good fit for your computer, trust that suggestion.
If Something Goes Wrong
- A model will not load or crashes: it is probably too big for your memory. Pick a smaller model or a more compressed version (a "Q4" one).
- Replies are slow: try a smaller model, and close other heavy apps to free memory.
- You cannot find a model in search: check your spelling, or try searching the model family name only, like "Gemma," rather than a long exact title.
Key Takeaways
- LM Studio (lmstudio.ai) is a free, fully graphical app for running local AI: no terminal at any point.
- Use the built-in catalog to search, download, load, and chat, all with your mouse.
- It is free for personal use and, per its makers, for commercial use too.
- Ollama and LM Studio are both free and private; LM Studio is the most point-and-click, Ollama is the leanest. Many people keep both.
- Model names with codes like "Q4" refer to compression, which the next lesson explains.