Ollama

Ollama is a free, open-source tool for running large language models such as Llama, Mistral, Gemma, Qwen and DeepSeek directly on a Mac, Windows or Linux computer. One command downloads a quantised model and starts a local chat or API server, so prompts, documents and code never leave your device. Developers use it for offline coding assistants, private document chat and cheap prototyping, while teams choose it when data privacy or air-gapped deployment matters more than frontier model power.