🦙

Get up and running with large language models locally.

Run Llama 3, Mistral, Gemma, and other open-weight models on your machine.

Run any model with a single command.

Ollama makes it incredibly easy to pull and run open-source models without complex configuration.

# Run Llama 3
ollama run llama3

Supported Apps

Cursor
OpenWebUI
Enchanted
Obsidian

Customize your own models.

Create customized models using your own parameters and system prompts via a simple Modelfile.

FROM llama3
PARAMETER temperature 1
SYSTEM """You are Mario from Super Mario Bros."""