How to install Ollama on Windows and run your first AI model
One command downloads and runs a model. Good if you are comfortable with a terminal, or want other apps and coding tools to use your local model. These steps use Qwen3 8B as the example; swap in any model.
Step by step
You need: Windows 10 (22H2) or Windows 11. For an NVIDIA card, a current driver from nvidia.com. Newer AMD Radeon cards (RX 6000 series and up) are supported.
- Go to ollama.com/download and download the Windows installer.
- Run OllamaSetup.exe. When it finishes, Ollama keeps running in the background (look for the llama icon by the clock).
- Open PowerShell (search for "PowerShell" in the Start menu) and run the command below. The first time, it downloads Qwen3 8B, which can take a while.
ollama run qwen3:8b- When you see
>>>, type a message and press Enter. Type/byeto quit. Ollama's own app window can also chat with any model you have downloaded.
That's it: you're running AI on your own computer. If something goes wrong, see the fixes below.
Not sure which model?
Answer two questions and get the model that fits your hardware, with these steps filled in for it.
Prefer LM Studio?
Easiest for beginners. A normal app with a chat window: you search for a model, click download, and start typing. No commands.
Which model to try first
The best everyday chat model that runs at a comfortable speed for each amount of graphics memory. Copy the command to download and run it.
| Your GPU | Model | Command |
|---|---|---|
| 8 GB graphics card |  Qwen3 8B | ollama run qwen3:8b |
| 12 GB graphics card |  Gemma 3 12B | ollama run gemma3:12b |
| 16 GB graphics card |  gpt-oss 20B | ollama run gpt-oss:20b |
| 24 GB graphics card | Qwen3 32B | ollama run qwen3:32b |
| 32 GB graphics card | Qwen3 32B | ollama run qwen3:32b |
If something goes wrong
The four problems beginners hit most, and how to fix them.
Out of memory or won't load
The model plus its working memory doesn't fit in your graphics memory. Here is how to make it fit, from the easiest fix to the most advanced.
It's really slow
Most slowness comes from the model not fitting fully in graphics memory, or from "thinking" models reasoning before they answer.
It's not using my GPU
If your graphics card sits idle while the CPU does all the work, one of these is usually the reason.
Weird, wrong or repeating answers
Local models can be excellent, but small or badly configured ones make mistakes that paid assistants don't. Here is what to check.
Frequently asked questions
Is Ollama free?
Yes. Ollama is free to download and use, and so are the models in this guide.
Does Ollama on Windows need the internet?
Only to download the app and the model. After that, chats run entirely on your computer and work offline.
How much space do I need for Ollama?
The app itself is small. Each model is a separate download, typically 2 to 20 GB; Qwen3 8B is about 5 GB.