As for poker, Google DeepMind selected heads-up no-Restrict Texas Keep’em as its benchmark for this experiment. Game Arena is working to be a heads-up poker tournament between major AI models, with outcomes feeding into a community leaderboard. Google DeepMind is expanding its Game Arena System to benchmark AI styles in https://cristianculct.wikicommunications.com/7138969/a_secret_weapon_for_game_arena