LocalAI

Why we included this project

LocalAI is a self-hosted inference server with an OpenAI-compatible API, so teams can point existing applications at local models without rewriting their code. It runs on CPU-only hardware and modest machines, which makes it a practical choice when cloud GPU costs or data-residency rules rule out hosted providers. The project puts llama.cpp, vLLM, whisper.cpp, stable-diffusion, and other engines behind a single interface, so one deployment can serve text, vision, speech, and image models. Backends are pulled on demand rather than bundled, so you only install what the models you actually use need. For developers and small teams that want a self-contained alternative to hosted APIs, the quickstart path and broad model support are the main draws.