Leaderboard
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.
| Rank | Model | Reasoning | Score | Programs | Raw rating | Combined record (W / D / L) | W / D / L % |
|---|---|---|---|---|---|---|---|
| 1 |  GPT-5.4 | Medium | 97.1 | 1 working of 1 attempt | 2291.1 | 510 / 8 / 10 | 97% / 2% / 2% |
| 2 | GPT-6 Astra | XHigh | 96.9 | 1 working of 1 attempt | 2278.7 | 513 / 0 / 15 | 97% / 0% / 3% |
| 3 | GPT-5.4 | XHigh | 94.8 | 1 working of 1 attempt | 2168.0 | 494 / 15 / 19 | 94% / 3% / 4% |
| 4 |  Gemini 4 Argon | High | 94.7 | 1 working of 1 attempt | 2162.9 | 497 / 0 / 27 | 95% / 0% / 5% |
| 5 |  Claude Opus 5.5 | Medium | 92.9 | 1 working of 1 attempt | 2097.9 | 472 / 35 / 19 | 90% / 7% / 4% |
| 6 | GPT-6 Astra | Low | 92.1 | 1 working of 1 attempt | 2071.1 | 473 / 24 / 29 | 90% / 5% / 6% |
| 7 | Claude Fable 5.1 | XHigh | 92.1 | 1 working of 1 attempt | 2070.5 | 482 / 2 / 40 | 92% / <1% / 8% |
| 8 |  Kimi K3.1 Pro | Max | 91.9 | 1 working of 1 attempt | 2064.0 | 487 / 1 / 42 | 92% / <1% / 8% |
| 9 | Gemini 4 Argon | Low | 91.2 | 1 working of 1 attempt | 2045.0 | 482 / 0 / 46 | 91% / 0% / 9% |
| 10 |  Muse Spark Open | XHigh | 90.9 | 1 working of 1 attempt | 2034.8 | 480 / 0 / 48 | 91% / 0% / 9% |
| 11 |  Pareto 26.10 | Default | 90.5 | 1 working of 1 attempt | 2025.5 | 480 / 0 / 50 | 91% / 0% / 9% |
| 12 | Gemini 3.5 Flash | High | 89.9 | 1 working of 1 attempt | 2011.1 | 461 / 21 / 42 | 88% / 4% / 8% |
| 13 | Gemini 4 Argon | Medium | 89.3 | 1 working of 1 attempt | 1995.2 | 468 / 0 / 56 | 89% / 0% / 11% |
| 14 | Claude Opus 5.5 | Low | 88.6 | 1 working of 1 attempt | 1979.4 | 457 / 22 / 49 | 87% / 4% / 9% |
| 15 | Claude Fable 5.1 | Low | 86.7 | 1 working of 1 attempt | 1938.4 | 454 / 4 / 68 | 86% / 1% / 13% |
| 16 | Muse Spark 1.2 | Medium | 84.8 | 1 working of 1 attempt | 1902.0 | 434 / 24 / 68 | 83% / 5% / 13% |
| 17 | Claude Fable 5 | XHigh | 84.4 | 1 working of 1 attempt | 1894.2 | 443 / 1 / 82 | 84% / <1% / 16% |
| 18 | Muse Wonder | XHigh | 83.8 | 1 working of 1 attempt | 1883.1 | 439 / 6 / 83 | 83% / 1% / 16% |
| 19 | Claude Opus 5.5 | XHigh | 82.7 | 1 working of 1 attempt | 1864.8 | 430 / 9 / 87 | 82% / 2% / 17% |
| 20 |  Grok 4.5 | High | 82.1 | 1 working of 1 attempt | 1855.9 | 377 / 109 / 40 | 72% / 21% / 8% |
| 21 |  Glimpse | Enabled | 82.0 | 1 working of 1 attempt | 1853.4 | 422 / 21 / 85 | 80% / 4% / 16% |
| 22 |  DeepSeek V4.1 Flash 1017 | High | 81.8 | 1 working of 1 attempt | 1851.0 | 430 / 0 / 96 | 82% / 0% / 18% |
| 23 | Grok 4.5 | Medium | 81.8 | 1 working of 1 attempt | 1850.6 | 406 / 48 / 72 | 77% / 9% / 14% |
| 24 | GPT-5.5 | XHigh | 79.5 | 1 working of 1 attempt | 1815.0 | 356 / 123 / 47 | 68% / 23% / 9% |
| 25 | GPT-5.6 Luna | XHigh | 79.4 | 1 working of 1 attempt | 1813.7 | 417 / 0 / 109 | 79% / 0% / 21% |
Underlying results
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.
267 player programs.
| Model | Reasoning | Status | Score | Rating | W / D / L | Matches | Opponents | Notes | Generated |
|---|---|---|---|---|---|---|---|---|---|
GPT-5.4E-d4fa7dbb9458c464 | Medium | Rated Interval validated | 97.1 | 2291 | 510 / 8 / 10 | 528 / 528 | 264 | 8 capped draws | 2026-08-09First try |
GPT-6 AstraE-fc5055140bf78475 | XHigh | Rated Interval validated | 96.9 | 2279 | 513 / 0 / 15 | 528 / 528 | 264 | — | 2026-09-05First try |
GPT-5.4E-9879a37f54e31cff | XHigh | Rated Interval validated | 94.8 | 2168 | 494 / 15 / 19 | 528 / 528 | 264 | 15 capped draws | 2026-08-09First try |
Gemini 4 ArgonE-c9b7bf650530d8ba | High | Rated Interval validated | 94.7 | 2163 | 497 / 0 / 27 | 524 / 524 | 262 | — | 2026-10-21First try |
Claude Opus 5.5E-9e6850818d33c592 | Medium | Rated Interval validated | 92.9 | 2098 | 472 / 35 / 19 | 526 / 526 | 263 | 35 capped draws | 2026-09-22First try |
Gemini 3.6 FlashE-7484f4e7d8863a3b | Medium | Rated Interval validated | 92.9 | 2097 | 482 / 11 / 31 | 524 / 524 | 262 | 11 capped draws | 2026-07-24First try |
GPT-6 AstraE-7fbf151da7860da4 | Low | Rated Interval validated | 92.1 | 2071 | 473 / 24 / 29 | 526 / 526 | 263 | 24 capped draws | 2026-09-05First try |
Claude Fable 5.1E-443622d1ef9c6f19 | XHigh | Rated Interval validated | 92.1 | 2071 | 482 / 2 / 40 | 524 / 524 | 262 | 2 capped draws | 2026-09-08First try |
Kimi K3.1 ProE-030c793cb8126246 | Max | Rated Interval validated | 91.9 | 2064 | 487 / 1 / 42 | 530 / 530 | 265 | 1 capped draw | 2026-10-24First try |
Gemini 4 ArgonE-b29fd1d6479f1bc8 | Low | Rated Interval validated | 91.2 | 2045 | 482 / 0 / 46 | 528 / 528 | 264 | — | 2026-10-21First try |
Claude Opus 5E-700f3dddd611df07 | XHigh | Rated Interval validated | 90.9 | 2035 | 479 / 6 / 45 | 530 / 530 | 265 | 6 capped draws | 2026-07-25First try |
Muse Spark OpenE-482f21724d9e9898 | XHigh | Rated Interval validated | 90.9 | 2035 | 480 / 0 / 48 | 528 / 528 | 264 | — | 2026-10-20First try |
Pareto 26.10E-74fb208ee7158901 | Provider | Rated Interval validated | 90.5 | 2025 | 480 / 0 / 50 | 530 / 530 | 265 | — | 2026-10-21First try |
Gemini 3.5 FlashE-1802f43944b4b6f0 | High | Rated Interval validated | 89.9 | 2011 | 461 / 21 / 42 | 524 / 524 | 262 | 21 capped draws | 2026-09-03First try |
Gemini 4 ArgonE-31ef2459454f5e17 | Medium | Rated Interval validated | 89.3 | 1995 | 468 / 0 / 56 | 524 / 524 | 262 | — | 2026-10-21First try |
Claude Opus 5E-865d68ab219fd7ee | None | Rated Interval validated | 89.0 | 1989 | 433 / 67 / 24 | 524 / 524 | 262 | 67 capped draws | 2026-07-25First try |
Claude Opus 5.5E-c6c2699b24183df3 | Low | Rated Interval validated | 88.6 | 1979 | 457 / 22 / 49 | 528 / 528 | 264 | 22 capped draws | 2026-09-22First try |
Claude Fable 5.1E-d04c4d8635b641f9 | Low | Rated Interval validated | 86.7 | 1938 | 454 / 4 / 68 | 526 / 526 | 263 | 4 capped draws | 2026-09-03First try |
Muse Spark 1.2E-052cdf1e5975b860 | Medium | Rated Interval validated | 84.8 | 1902 | 434 / 24 / 68 | 526 / 526 | 263 | 24 capped draws | 2026-08-13First try |
Claude Fable 5E-99c76fc062487714 | XHigh | Rated Interval validated | 84.4 | 1894 | 443 / 1 / 82 | 526 / 526 | 263 | 1 capped draw | 2026-09-03First try |
Muse WonderE-5c48ba07caee5554 | XHigh | Rated Interval validated | 83.8 | 1883 | 439 / 6 / 83 | 528 / 528 | 264 | 6 capped draws | 2026-10-12First try |
Claude Opus 5.5E-05f63acd714d31e5 | XHigh | Rated Interval validated | 82.7 | 1865 | 430 / 9 / 87 | 526 / 526 | 263 | 9 capped draws | 2026-09-22First try |
Grok 4.5E-1c8e452ede41ea2e | High | Rated Interval validated | 82.1 | 1856 | 377 / 109 / 40 | 526 / 526 | 263 | 109 capped draws | 2026-07-24Repaired |
GlimpseE-160753088cb3a492 | Enabled | Rated Interval validated | 82.0 | 1853 | 422 / 21 / 85 | 528 / 528 | 264 | 21 capped draws | 2026-10-21First try |
DeepSeek V4.1 Flash 1017E-5074fe2d98bd9ce2 | High | Rated Interval validated | 81.8 | 1851 | 430 / 0 / 96 | 526 / 526 | 263 | — | 2026-10-20First try |