Get up and running with large language models.
Run Llama 4, GPT-OSS, DeepSeek, and more — locally, on your machine. Or scale to the cloud when you need to.
Quickstart
Three commands. You're running.
Install, pull a model, start chatting — works the same on every platform.
curl -fsSL https://ollama.com/install.sh | sh
ollama pull gpt-oss:120b
ollama run gpt-oss:120b
Library
See all
Featured models
Why Ollama
Local first. Cloud when you need it.
Runs on your machine
Apple Silicon, NVIDIA, AMD. No data leaves your device — full conversations stay local.
Cloud when you need scale
Spill over to Ollama Cloud for models that don't fit on your laptop. Same CLI, same API.
Drop-in OpenAI API
Compatible with the OpenAI API surface. Point your client at localhost:11434 and ship.
Ready when you are.
Open source, free for personal use. No telemetry. No login required to run locally.