LocalAI

LocalAI is an open-source, self-hosted AI runtime that serves as a local alternative to cloud-based AI APIs like OpenAI, Anthropic, and others. It is designed to run a wide variety of AI models—including text, vision, speech, audio, images, video, embeddings, reranking, and autonomous agents—on a single modular stack. The platform supports deployment from a CPU laptop to a distributed GPU cluster, giving users full control over their data and infrastructure. LocalAI integrates over 60 backends, such as llama.cpp, vLLM, SGLang, MLX, whisper.cpp, and diffusion engines, which are pulled on demand to keep the core lean. The project is MIT licensed and encourages community contributions.