Ollama
Run open LLMs locally with one command.
What is Ollama?
Ollama lets you download and run open-weight language models on your own machine.
It exposes a simple local API so apps can talk to models without the cloud.
Ollama is for running open-weight models on your own machine with one command. It suits developers who want to experiment without per-request costs, anyone whose data must not leave the device, and apps that need a local model behind a simple API.
What you can run depends on your hardware: larger models need a lot of memory, and smaller quantised models trade some quality for speed. There is no managed hosting, so keeping it running in production is your responsibility.
If you do not have capable hardware, a hosted API is simpler. For a chat interface on top of local models, Open WebUI pairs with Ollama; for a full self-hosted agent, SelfAgent.
Installation guide
Step 1 — Download the installer from ollama.com for your OS.
Step 2 — Run ollama run llama3 to pull and start a model.
Pros
- Free and open source
- Runs fully offline
- Simple CLI
Cons
- Needs a capable machine
- No managed hosting
Frequently asked questions
What is Ollama?
Run open LLMs locally with one command.
Is Ollama free?
Ollama is open source and free to use.
What are alternatives to Ollama?
Check our AI tools directory for alternatives to Ollama.
What platforms does Ollama support?
Ollama is available on macOS, Windows, Linux.