mudler/LocalAI

LocalAI is a self-hosted, open-source alternative to proprietary AI APIs, providing a unified gRPC and HTTP interface for running diverse AI models locally. The system orchestrates a modular backend architecture that supports text generation, speech-to-text, text-to-speech, image and video generation, and various computer vision tasks. It features a comprehensive web UI for model management, a distributed mode for scaling across nodes, and compatibility layers for OpenAI, Ollama, and other standard APIs.

This period focused on significantly expanding the backend ecosystem by introducing support for Python-based inference frameworks (vLLM, Diffusers, Transformers, Coqui) and C++ audio processing, alongside a major architectural shift to gRPC-based backend communication. Concurrently, the project modernized its web interface with a React-based UI, added agent job management, and implemented federated server capabilities for distributed inference.

55 changes

This period focused on significantly expanding the native Go and C++ backend ecosystem, introducing numerous new backends for speech, image, and 3D generation while replacing Python dependencies with high-performance alternatives. Simultaneously, the project underwent a major frontend migration to a standalone React application, introducing a modern, component-based architecture with advanced state management and internationalization.

116 changes

mudler/LocalAI was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point the surveyor at a repository you know and see whether you agree with it.