from Hacker News

I was inspired by https://news.ycombinator.com/item?id=43586380 to simulate the reality TV show "The Traitors" using open-source LLMs. I used all open source via either groq or locally hosted on llama-server (llama.cpp)

Watching the games unfold is pretty entertaining, even when the very small models lose track of the plot.(like a outmatched severely quantized phi4)

I plan to improve the visualizations over time.

The repo is here: https://github.com/michaelgiba/the-traitors

One game's visualization is here: https://michaelgiba.github.io/the-traitors/results/traitors/...