LocalAI
ActiveDescription
Open-source AI engine to run any model — LLMs, vision, voice, image, video — on any hardware without GPU. Provides OpenAI-compatible API for fully local, privacy-first AI inference.
Key Features
- Composable backend architecture — Each backend (llama.cpp, vLLM, whisper.cpp, etc.) is a separate image pulled on demand
- OpenAI/Anthropic/ElevenLabs-compatible API — Unified API interface across all backends for seamless cloud replacement
- Any modality support — LLMs, vision, voice, image, and video models behind a single unified API
- Any hardware support — Runs on NVIDIA, AMD, Intel, Apple Silicon, Vulkan, or CPU-only configurations
- Built-in AI agents — Autonomous agents with tool use, RAG, MCP, and skills support
- Multi-user ready — API key authentication, per-user quotas, and role-based access control
Use Cases
💡 Private AI infrastructure: Build fully localized AI inference platforms for enterprises with data staying on-premise
💡 Multimodal AI application development: Access LLMs, speech recognition, image generation through a unified API
💡 Multi-tenant AI platforms: Build user-facing AI services with API key and quota management
💡 Edge device deployment: Run AI models on GPU-less edge devices for industrial IoT and embedded scenarios
Categories
Quick Start
# One-click Docker start (CPU only)
docker run -ti --name local-ai -p 8080:8080 localai/localai:latest
# NVIDIA GPU start
docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-12
# Load model and run
local-ai run llama-3.2-1b-instruct:q4_k_m
# Load from HuggingFace
local-ai run huggingface://TheBloke/phi-2-GGUF/phi-2.Q8_0.gguf