Hey folks, I wanted to share a project I’ve been working on called Weighthouse.
A while back I started to hit my mental limit when it came to managing my local model collection by hand. I found and started using HuggingHack to manage them. I liked it, but my setup had grown beyond a single storage location, and I wanted it to do more than download and archive models.
I wanted it to answer questions like: which revision did I actually download? Are all the files still there? What changed upstream since I downloaded my copy? How do I move a model between storage locations without managing everything by hand?
So I forked HuggingHack, started extending it for my own environment, and eventually renamed the project Weighthouse.
Weighthouse is a self-hosted model library. It manages Hugging Face repositories on storage you control, with a web interface for people and an MCP server and API for agents.
A few of the things it does:
Updates aren’t automatic, and files removed upstream aren’t silently deleted from local storage.
The default installation runs in one container with SQLite, with PostgreSQL available if you prefer an external database. It’s MIT licensed.
I’ve been using it to manage my own collection, but I’d be interested in how other people handle model revisions and storage across multiple machines or storage pools. I’m particularly interested in workflows I’ve missed.
Project: [https://weighthouse.ai](https://weighthouse.ai/)
Source: [https://github.com/weighthouse/weighthouse](https://github.com/weighthouse/weighthouse)
And a big thanks to tyedalwaves for releasing HuggingHack in the first place. This project wouldn’t exist without that starting point. (And my apologies for not linking directly to HuggingHack or tyedalwaves, as a new user here I’m limited to 2 links in a post).