LocalAI

mudler/LocalAI

Run any AI model on any hardware, no GPU required

Why pick itSupports multiple models and hardware, flexible and privacy-focused

Install

Requires Docker environment, supports multiple hardware and models.

Summarized by AI from the project docs; check the official docs before relying on it.

rtkRun locally · ★ 85k
CLI proxy that reduces LLM token consumption by 60-90%
unslothRun locally · ★ 78.8k
GUI app to run and train large language models and diffusion models locally
private-gptRun locally · ★ 57.6k
Complete API layer for private AI applications on local models: RAG, skills, tools, MCP, and more
codebase-memory-mcpRun locally · ★ 47.8k
Local code intelligence tool that connects code, docs, and infrastructure in a persistent graph.
OllamaRun locally · ★ 185k
Download and run open-weight models like Qwen, DeepSeek, Gemma and gpt-oss locally with one command, with an OpenAI-compatible API.