Ollama

Lightweight framework for running open-source LLMs locally with a single terminal command. Supports Llama, Mistral, Gemma, Qwen, and 100+ models. Exposes an OpenAI-compatible API so any app built for OpenAI can point to your local models instead. Works on Mac, Windows, and Linux. Completely free and open-source under MIT license. Best for developers and privacy-conscious users who want full control over their AI stack without cloud costs or data exposure.

Browse tasks in this category · Category overview