LocalAI is an open-source server for running AI models on your own hardware, with an API that follows the OpenAI and Anthropic formats. Point your client's base URL at http://localhost:8080/v1 and existing code works.
It covers chat, vision, image, video and music generation, speech-to-text, text-to-speech, embeddings and reranking, with GPU acceleration on NVIDIA, AMD, Intel and Apple hardware or CPU only. Development builds add a file workspace, a Python sandbox and multi-user logins.
Install with curl -sSL https://localai.io/install.sh | sh, the localai/localai:latest Docker image, a Helm chart, or the desktop app for Linux and macOS. MIT licensed and free.