hey everyone, Ken here. you can lose an evening to READMEs and star counts picking one open-source AI tool out of a few dozen. I put the facts for 322 of them in one place: chat UIs, model servers, agents, coding agents, Claude Code skills, starter templates. each project gets a score out of 100 from public GitHub data (adoption, last commit and release, how issues get handled, how easy it is to install), a review with what it does well and where it gets annoying, and its license in plain words. you can search with filters, open 170 head-to-head pages like llama.cpp vs ollama, and pull everything from an open JSON API under CC BY 4.0. what it can't tell you yet: whether a project builds and starts (no tested builds so far), RAM for most apps, and trends over time, which turn on once there are 30 days of history (around november 7). nobody pays for a spot and I sell none of the tools. I do build AI apps for companies, and that never moves a rank; the about page has the rules. if a fact about a project you know is wrong, tell me here and I'll fix it.

Archestack

Pick open-source AI tools with the facts on the table

Pick open-source AI tools with the facts on the table