What it does
This game playground compares Jev with other evaluators using explicit states, legal moves, and observable outcomes.
Benchmarks & research
Screenshot unavailable. Open the experiment ↗
This game playground compares Jev with other evaluators using explicit states, legal moves, and observable outcomes.
You can compare how different artificial intelligence models make tactical decisions in a turn-based board game. Start by sketching out your game rules and writing down every permitted move for each board position. A developer can build the game referee so it only presents these valid options to the models.
Your team can watch each model pick from the exact same list of choices across separate boards. Keep in mind that this playground only lets you inspect individual moves one turn at a time. It does not run automated tournaments across many completed matches to calculate overall win rates.