As for poker, Google DeepMind decided on heads-up no-Restrict Texas Maintain’em as its benchmark for this experiment. Game Arena is running like a heads-up poker tournament between top AI models, with results feeding into a general public leaderboard. Google DeepMind is increasing its Game Arena System to benchmark AI types https://johnnyjzmyj.hamachiwiki.com/2217876/game_arena_an_overview