By model and setting

Each model and reasoning setting appears once. If a model produced more than one player program for this game, the row stands for all of them: Programs counts the working attempts out of the counted attempts, Score and Raw rating average the programs (weighted by attempts; the small range is the lowest and highest program rating), and the combined record adds up their matches. Score is relative within this game (0–100). Scores here are working-program strength: failed generation attempts are left out and count as 0 only in a model's headline score.

GB2-01: Go - Reasoning Variants
Rank Model Reasoning Score Programs Raw rating Combined record (W / D / L)
1GPT-6.2 AstraXHigh99.31 working of 1 attempt2512.1674 / 0 / 6
2Claude Fable 5.2XHigh98.11 working of 1 attempt2322.1661 / 0 / 19
3Gemini 4 CarbonHigh96.61 working of 1 attempt2205.5645 / 3 / 32
4GPT-6.2 Sol ProXHigh96.61 working of 1 attempt2202.2642 / 8 / 30
5GPT-6.2 Sol ProMedium96.41 working of 1 attempt2192.7633 / 23 / 24
6GPT-5.4Medium96.31 working of 1 attempt2183.4635 / 16 / 29
7GPT-6 AstraXHigh96.31 working of 1 attempt2183.4643 / 0 / 37
8Gemini 4 CarbonLow96.21 working of 1 attempt2177.4642 / 0 / 38
9GPT-5.4XHigh94.51 working of 1 attempt2099.1615 / 24 / 41
10GPT-6.2 Astra ProXHigh94.51 working of 1 attempt2099.1602 / 50 / 28
11GPT-6 Astra ProXHigh94.31 working of 1 attempt2090.0618 / 14 / 48
12Gemini 4 ArgonHigh94.11 working of 1 attempt2085.5624 / 0 / 56
13Gemini 4 NeonXHigh93.81 working of 1 attempt2072.4621 / 0 / 59
14GPT-6.2 SolMedium93.81 working of 1 attempt2072.4616 / 10 / 54
15Claude Fable 5.2Medium93.11 working of 1 attempt2047.7615 / 0 / 65
16Gemini 4 HydrogenXHigh92.51 working of 1 attempt2028.4610 / 0 / 70
17Claude Opus 5.5Medium92.41 working of 1 attempt2026.5587 / 45 / 48
18Claude Fable 5.1XHigh91.81 working of 1 attempt2006.5603 / 2 / 75
19Gemini 4 NeonMedium91.81 working of 1 attempt2006.5604 / 0 / 76
20GPT-6 AstraLow91.81 working of 1 attempt2006.5591 / 26 / 63
21Kimi K3.1 ProMax91.31 working of 1 attempt1992.6599 / 2 / 79
22Particle-1Max91.01 working of 1 attempt1984.1582 / 31 / 67
23Ernie 5.5 Flash PreviewDefault90.61 working of 1 attempt1975.9580 / 30 / 70
24Gemini 4 ArgonLow90.51 working of 1 attempt1972.6594 / 0 / 86
25GPT-6 Astra ProMedium90.11 working of 1 attempt1963.0570 / 42 / 68

Individual player programs

Each row is one accepted player program for GB2-01: Go; a model can have more than one. Score is the 0–100 result in this game, and its rating and W / D / L record show how it played. Scores are working-program strength: failed generation attempts have no row and count only in a model's headline score.

339 player programs.

ModelReasoningStatusScoreRatingW / D / LMatchesOpponentsNotesGenerated
GPT-6.2 AstraE-ded6677fc4d2c63cXHighRated Interval validated99.32512674 / 0 / 6680 / 680340—2026-11-17First try
Claude Fable 5.2E-8ee35c299845147fXHighRated Interval validated98.12322661 / 0 / 19680 / 680340—2026-11-09First try
Gemini 4 CarbonE-8f7c0faa1d26ca79HighRated Interval validated96.62205645 / 3 / 32680 / 6803403 capped draws2026-11-12First try
GPT-6.2 Sol ProE-86f4db4fea47e907XHighRated Interval validated96.62202642 / 8 / 30680 / 6803408 capped draws2026-11-18First try
GPT-6.2 Sol ProE-b29baf67617e0fd0MediumRated Interval validated96.42193633 / 23 / 24680 / 68034023 capped draws2026-11-18First try
GPT-5.4E-d4fa7dbb9458c464MediumRated Interval validated96.32183635 / 16 / 29680 / 68034016 capped draws2026-08-09First try
GPT-6 AstraE-fc5055140bf78475XHighRated Interval validated96.32183643 / 0 / 37680 / 680340—2026-09-05First try
Gemini 4 CarbonE-3dddac17b1d0d79fLowRated Interval validated96.22177642 / 0 / 38680 / 680340—2026-11-12First try
GPT-5.4E-9879a37f54e31cffXHighRated Interval validated94.52099615 / 24 / 41680 / 68034024 capped draws2026-08-09First try
GPT-6.2 Astra ProE-85dd76f643c70002XHighRated Interval validated94.52099602 / 50 / 28680 / 68034050 capped draws2026-11-18First try
GPT-6 Astra ProE-70601c965e785978XHighRated Interval validated94.32090618 / 14 / 48680 / 68034014 capped draws2026-11-17First try
Gemini 4 ArgonE-c9b7bf650530d8baHighRated Interval validated94.12085624 / 0 / 56680 / 680340—2026-10-21First try
Gemini 3.6 FlashE-7484f4e7d8863a3bMediumRated Interval validated93.92077612 / 20 / 48680 / 68034020 capped draws2026-07-24First try
Gemini 4 NeonE-adf80bd708c51b26XHighRated Interval validated93.82072621 / 0 / 59680 / 680340—2026-11-19First try
GPT-6.2 SolE-02f27435ee4f7f39MediumRated Interval validated93.82072616 / 10 / 54680 / 68034010 capped draws2026-11-17First try
Claude Fable 5.2E-54829f6a69e69c0fMediumRated Interval validated93.12048615 / 0 / 65680 / 680340—2026-11-09First try
Gemini 4 HydrogenE-9e6063f70f796448XHighRated Interval validated92.52028610 / 0 / 70680 / 680340—2026-11-12First try
Claude Opus 5.5E-9e6850818d33c592MediumRated Interval validated92.42026587 / 45 / 48680 / 68034045 capped draws2026-09-22First try
Claude Fable 5.1E-443622d1ef9c6f19XHighRated Interval validated91.82006603 / 2 / 75680 / 6803402 capped draws2026-09-08First try
Gemini 4 NeonE-837ab62b370fb059MediumRated Interval validated91.82006604 / 0 / 76680 / 680340—2026-11-19First try
GPT-6 AstraE-7fbf151da7860da4LowRated Interval validated91.82006591 / 26 / 63680 / 68034026 capped draws2026-09-05First try
Kimi K3.1 ProE-030c793cb8126246MaxRated Interval validated91.31993599 / 2 / 79680 / 6803402 capped draws2026-10-24First try
Claude Opus 5E-700f3dddd611df07XHighRated Interval validated91.01986595 / 6 / 79680 / 6803406 capped draws2026-07-25First try
Particle-1E-2d3bce96a1836811MaxRated Interval validated91.01984582 / 31 / 67680 / 68034031 capped draws2026-11-03First try
Ernie 5.5 Flash PreviewE-9d9386c86e5364b7DefaultRated Interval validated90.61976580 / 30 / 70680 / 68034030 capped draws2026-11-12First try