CoupBench

83 games of Coup between four LLMs, with every hidden card shown. Run the benchmark / contribute

GameWinnerTurns
Game 1Claude Opus 5.59
Game 2MiMo V2.6 Pro9
Game 3Claude Opus 5.59
Game 4MiMo V2.6 Pro13
Game 5GPT-6.1 Sol10
Game 6GPT-6.1 Sol12
Game 7MiMo V2.6 Pro10
Game 8GPT-6.1 Sol11
Game 9GPT-6.1 Sol14
Game 10GPT-6.1 Sol18
Game 11MiMo V2.6 Pro9
Game 12MiMo V2.6 Pro13
Game 13Claude Opus 5.59
Game 14GPT-6.1 Sol14
Game 15GPT-6.1 Sol15
Game 16Claude Opus 5.58
Game 17Mistral Large 48
Game 18MiMo V2.6 Pro9
Game 19GPT-6.1 Sol14
Game 20GPT-6.1 Sol11
Game 21MiMo V2.6 Pro13
Game 22MiMo V2.6 Pro9
Game 23Claude Opus 5.513
Game 24Claude Opus 5.510
Game 25Claude Opus 5.58
Game 26Claude Opus 5.517
Game 27Claude Opus 5.511
Game 28Mistral Large 421
Game 29Claude Opus 5.514
Game 30MiMo V2.6 Pro12
Game 31Claude Opus 5.511
Game 32MiMo V2.6 Pro12
Game 33Claude Opus 5.512
Game 34GPT-6.1 Sol7
Game 35Claude Opus 5.59
Game 36Claude Opus 5.59
Game 37MiMo V2.6 Pro12
Game 38MiMo V2.6 Pro15
Game 39Mistral Large 412
Game 40Claude Opus 5.510
Game 41MiMo V2.6 Pro11
Game 42MiMo V2.6 Pro10
Game 43MiMo V2.6 Pro13
Game 44Mistral Large 415
Game 45Mistral Large 411
Game 46GPT-6.1 Sol9
Game 47GPT-6.1 Sol19
Game 48Mistral Large 410
Game 49Claude Opus 5.511
Game 50MiMo V2.6 Pro16
Game 51Claude Opus 5.518
Game 52GPT-6.1 Sol22
Game 53MiMo V2.6 Pro10
Game 54MiMo V2.6 Pro13
Game 55Mistral Large 49
Game 56Mistral Large 410
Game 57Claude Opus 5.518
Game 58GPT-6.1 Sol12
Game 59Claude Opus 5.513
Game 60GPT-6.1 Sol10
Game 61GPT-6.1 Sol12
Game 62GPT-6.1 Sol11
Game 63GPT-6.1 Sol13
Game 64GPT-6.1 Sol11
Game 65GPT-6.1 Sol21
Game 66MiMo V2.6 Pro10
Game 67MiMo V2.6 Pro9
Game 68Mistral Large 49
Game 69MiMo V2.6 Pro5
Game 70Claude Opus 5.510
Game 71Claude Opus 5.58
Game 72MiMo V2.6 Pro13
Game 73Claude Opus 5.515
Game 74MiMo V2.6 Pro13
Game 75GPT-6.1 Sol13
Game 76GPT-6.1 Sol15
Game 77MiMo V2.6 Pro14
Game 78Claude Opus 5.513
Game 79GPT-6.1 Sol26
Game 80Mistral Large 413
Game 81GPT-6.1 Sol17
Game 82MiMo V2.6 Pro16
Game 83Claude Opus 5.511