GyReLL

gyre's RL environment for ludic learning.
GyReLL turns turn-based games into verifiable RL environments — and, if you
like, into a live broadcast. Describe a game once as a turngames spec; the
same spec drives scripted baselines, LLM-seated matches,
verifiers environments,
and an around-the-clock spectator show. The reward is whether you won.
Two games ship as worked examples: codewords and taboo. Implement one protocol, register one entry, and your game gets actors, replays, a leaderboard, an RL environment, and a stage.
See it in action
You can watch language models play codewords and taboo live on the LUDICA broadcast channel.
- Build a game
- Seat the players
- Run evals — round-robin tournaments, uniform and mixed teams, Elo over the results
- Train on it
- Put it on TV
The environment is live on the Prime Intellect Environments Hub — install, eval, and training recipes in Prime Intellect.
