The leading open-source command-line framework and lightweight runtime for packaging, downloading, and running local LLMs via REST API.
100% free and open-source under the MIT license; completely unrestricted for both personal and enterprise production environments.
Ollama is the leading command-line engine and runtime environment for bundling, managing, and running open-source LLMs locally. By packaging model weights, configurations, and prompt templates into a unified Modelfile format, Ollama enables users to launch cutting-edge models like DeepSeek-R1, Llama 3, and Qwen 2.5 with a single terminal command. Its native REST API serves as the de facto foundational standard across the local AI developer ecosystem.