LocalAI
Run any AI model without GPU required
45.4k|stable|v4.1.3
Multi-model inference — Run LLMs, vision, voice, image and video models on single instance.
Hardware flexibility — CPU-only or accelerated via NVIDIA, AMD, Intel, Apple Silicon, Vulkan.
API compatibility — Drop-in OpenAI, Anthropic, ElevenLabs API compatibility.
AI agents — Autonomous agents with tool use, RAG, MCP and skills.
Multi-user support — Built-in auth, quotas, and role-based access control.
LocalAI runs any AI model locally without GPU—perfect for privacy-first deployments. Supports 36+ backends and works on any hardware from CPUs to GPUs. Includes built-in agents, multi-user auth, and OpenAI-compatible APIs so you can drop it into existing workflows.
Docker or native binary installation; no GPU required but GPU acceleration available
