DEV Community

GDS K S
GDS K S

Posted on

I built a self-hosted AI list that checks itself weekly

Self-hosted AI moves fast. A list that was right in March has dead links by June, and nobody notices until they try to install something.

So I built a list that checks itself. awesome-selfhosted-ai has 166 open-source projects you can run on your own hardware: local LLM runtimes, chat UIs, agents, MCP servers, RAG and vector databases, speech, image generation, document OCR, voice agents and more.

GitHub logo glincker / awesome-selfhosted-ai

Curated list of self-hosted open-source AI: local LLMs, chat UIs, agents, MCP servers, RAG, speech and image generation. Health-checked weekly, dead projects removed automatically.

Awesome Self-Hosted AI Awesome

Open-source AI you can run on your own hardware or server.

166 projects across local LLMs, chat interfaces, agents, MCP servers, RAG, speech and image generation. A weekly job checks every project against the GitHub API, and anything archived, deleted or silent for over a year moves to UNMAINTAINED.md.

166 projects | 166 active in the last 90 days | 6.2M combined stars | 0 unmaintained | refreshed 2026-10-31

See the live health table for stars, last push and license per project.

Contents

How it stays current

A weekly GitHub Action reads every project's repo through the GitHub API and records stars, last push, license and status. If a project is archived, deleted, or has had no push for a year, it moves to a separate Unmaintained file on its own. If it comes back to life, it comes back to the list. Nothing depends on me remembering to prune.

What is in it

16 categories, from LLM runtimes to voice agents and home AI.
A live table with stars, last push, license and a colored status badge per project.
A source-available section for popular tools whose license adds terms beyond a standard open-source one (n8n, Open WebUI, Dify and similar). The license is shown, so you can check it before commercial use.
Rising projects that are under a year old, which older lists tend to miss.
Rules I kept

Open source (or clearly labeled source-available), self-hostable, not archived, pushed to within 90 days, at least 500 stars. One project per pull request, and the README is generated from one data file, so a pull request is a small JSON change that CI validates.

Honest note

I used AI tools to help find candidates and draft descriptions. A script verifies every repo against the GitHub API, and I review the list. If you spot a wrong license or a dead project, open an issue and it gets fixed.

Link: https://github.com/glincker/awesome-selfhosted-ai

What is missing? Tell me which project should be in it.

Top comments (0)