mudlerLocalAI
Local runtime for LLMs, vision, voice, image, and video models. No GPU required for many models.
TL;DR · 30-second scan
LocalAI (Go) — Local runtime for LLMs, vision, voice, image, and video models. No GPU required for many models.
You want one runtime for every open model type with OpenAI-compatible endpoints.
Voice & Audio Generation · AI Video Generation
The Swiss-army-knife for running open models locally. Not the fastest or most specialized runtime, but it handles LLMs, TTS, image, and video models under one API. Big win: OpenAI-compatible API surface, so any tool built for OpenAI just points at your localhost and works. If you are worried about API pricing, this is the escape hatch. MIT licensed, actively developed, runs on Docker or bare metal.
You want one runtime for every open model type with OpenAI-compatible endpoints.
You need production-grade throughput — dedicated runtimes (vLLM, TensorRT) beat this for LLM serving at scale.
Add this badge to your README to show your project is curated on StackPicks. Free, lightweight (180×28 SVG), and gives your visitors a one-click way to see honest take + alternatives.
[](https://stackpicks.dev/repo/localai)
<a href="https://stackpicks.dev/repo/localai"><img src="https://stackpicks.dev/api/badge/localai" alt="Featured on StackPicks" width="180" height="28" /></a>
Are you the maintainer of mudler/LocalAI? Add the badge and we'll feature your project in the weekly curator newsletter.