Ollama

About Ollama

Ollama makes running local LLMs on your Mac as easy as a single terminal command. Pull models like Llama 3, Mistral, Phi-3, and Gemma, then chat with them via a simple REST API or the Ollama CLI. No internet connection required once downloaded, no privacy concerns, and no API fees. Supports Apple Silicon acceleration via Metal. The de facto standard for local LLM experimentation on Mac — integrates with Open WebUI, Continue, and dozens of other clients.