Six AI decision models face a reproducible Pac-Man test
Six decision models now have a like-for-like Pac-Man benchmark, giving builders a checkable way to compare models that choose actions rather than generate text. Each model played 100 games against the same scripted ghosts, receiving the maze, pellets, and ghosts as JSON at every junction and returning probabilities for each direction. Answers taking over two seconds trigger a backup rule, and every game is recorded for replay.