ollama / ollama

Runs open-weight LLMs such as Gemma, Qwen, DeepSeek, and gpt-oss locally or in the cloud, with a CLI, REST API, and Python and JavaScript libraries.

mudler / LocalAI

Open-source local AI engine that runs LLMs, vision, voice, image, and video models behind OpenAI-compatible APIs on CPU or GPU.