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) |
|---|---|---|---|---|---|---|
| 1 |  GPT-6 Astra | XHigh | 98.9 | 1 working of 1 attempt | 2940.5 | 527 / 0 / 3 |
| 2 |  Claude Opus 5.5 | XHigh | 98.2 | 1 working of 1 attempt | 2816.1 | 523 / 0 / 7 |
| 3 | GPT-6.1 Sol | XHigh | 97.9 | 1 working of 1 attempt | 2767.5 | 521 / 0 / 9 |
| 4 | Claude Sonnet 5.5 | XHigh | 97.4 | 1 working of 1 attempt | 2705.3 | 518 / 0 / 12 |
| 5 |  Gemini 4 Argon | High | 96.3 | 1 working of 1 attempt | 2606.2 | 512 / 0 / 18 |
| 6 | GPT-6.1 Sol | Medium | 94.8 | 1 working of 1 attempt | 2493.6 | 503 / 0 / 27 |
| 7 | GPT-6 Astra | Low | 94.3 | 1 working of 1 attempt | 2461.9 | 500 / 0 / 30 |
| 8 |  Pareto 26.10 Preview | Provider | 94.3 | 1 working of 1 attempt | 2461.9 | 500 / 0 / 30 |
| 9 | GPT-6 Sol | XHigh | 93.8 | 1 working of 1 attempt | 2432.2 | 497 / 0 / 33 |
| 10 | Claude Fable 5.1 | XHigh | 93.4 | 1 working of 1 attempt | 2413.3 | 495 / 0 / 35 |
| 11 |  Grok 4.7 | XHigh | 93.4 | 1 working of 1 attempt | 2413.3 | 495 / 0 / 35 |
| 12 | Claude Opus 5 | XHigh | 93.3 | 4 | 2565.2 | 1976 / 0 / 144 |
| 13 | GPT-6 Astra | Medium | 93.1 | 1 working of 1 attempt | 2395.1 | 493 / 0 / 37 |
| 14 |  Muse Wonder | XHigh | 93.1 | 1 working of 1 attempt | 2395.1 | 493 / 0 / 37 |
| 15 | Grok 4.7 | Medium | 93.1 | 1 working of 1 attempt | 2395.1 | 493 / 0 / 37 |
| 16 |  Qwen3.8 Max 0902 | XHigh | 92.4 | 1 working of 1 attempt | 2360.3 | 489 / 0 / 41 |
| 17 | Claude Opus 5.5 | Low | 91.9 | 1 working of 1 attempt | 2335.4 | 486 / 0 / 44 |
| 18 | Pareto 26.10 | Provider | 91.9 | 1 working of 1 attempt | 2335.4 | 486 / 0 / 44 |
| 19 | Claude Fable 5.1 | Low | 91.7 | 1 working of 1 attempt | 2327.3 | 485 / 0 / 45 |
| 20 | GPT-6 Sol | Medium | 91.6 | 1 working of 1 attempt | 2319.4 | 484 / 0 / 46 |
| 21 | GPT-6.1 Sol | Low | 91.4 | 1 working of 1 attempt | 2311.5 | 483 / 0 / 47 |
| 22 | Qwen3.8 Max | Minimal | 90.6 | 1 working of 1 attempt | 2273.4 | 478 / 0 / 52 |
| 23 | Gemini 4 Argon | Medium | 90.4 | 1 working of 1 attempt | 2266.0 | 477 / 0 / 53 |
| 24 |  Kimi K3.1 Pro | Max | 90.4 | 1 working of 1 attempt | 2266.0 | 477 / 0 / 53 |
| 25 |  GLM-5.3 | Max | 90.2 | 1 working of 1 attempt | 2258.6 | 476 / 0 / 54 |
Underlying results
Individual player programs
Each row is one accepted player program for GB2-05: Hex; 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.
266 player programs.
| Model | Reasoning | Status | Score | Rating | W / D / L | Matches | Opponents | Notes | Generated |
|---|---|---|---|---|---|---|---|---|---|
Claude Opus 5E-3f3686d02d1503fa | XHigh | Rated Interval validated | 99.1 | 2981 | 528 / 0 / 2 | 530 / 530 | 265 | — | 2026-07-25First try |
GPT-6 AstraE-84a392f4c3b1b8df | XHigh | Rated Interval validated | 98.9 | 2940 | 527 / 0 / 3 | 530 / 530 | 265 | — | 2026-09-05First try |
Claude Opus 5.5E-17a0ec0d72fab9d9 | XHigh | Rated Interval validated | 98.2 | 2816 | 523 / 0 / 7 | 530 / 530 | 265 | — | 2026-09-22Repaired |
GPT-6.1 SolE-84e2760b1fa4b2a3 | XHigh | Rated Interval validated | 97.9 | 2767 | 521 / 0 / 9 | 530 / 530 | 265 | — | 2026-10-01First try |
Claude Sonnet 5.5E-8e760134a0746d5c | XHigh | Rated Interval validated | 97.4 | 2705 | 518 / 0 / 12 | 530 / 530 | 265 | — | 2026-09-30First try |
Gemini 4 ArgonE-520d168a984eb89c | High | Rated Interval validated | 96.3 | 2606 | 512 / 0 / 18 | 530 / 530 | 265 | — | 2026-10-21First try |
GPT-6.1 SolE-b18d2c85adab44db | Medium | Rated Interval validated | 94.8 | 2494 | 503 / 0 / 27 | 530 / 530 | 265 | — | 2026-10-01First try |
GPT-6 AstraE-bcfaa85bf834db66 | Low | Rated Interval validated | 94.3 | 2462 | 500 / 0 / 30 | 530 / 530 | 265 | — | 2026-09-05First try |
Pareto 26.10 PreviewE-938c7886323da9c0 | Provider | Rated Interval validated | 94.3 | 2462 | 500 / 0 / 30 | 530 / 530 | 265 | — | 2026-10-09First try |
GPT-6 SolE-5fda0fd12201228d | XHigh | Rated Interval validated | 93.8 | 2432 | 497 / 0 / 33 | 530 / 530 | 265 | — | 2026-09-22First try |
Claude Fable 5.1E-f1eb379197580e01 | XHigh | Rated Interval validated | 93.4 | 2413 | 495 / 0 / 35 | 530 / 530 | 265 | — | 2026-09-08Repaired |
Grok 4.7E-be483c51fbcfb505 | XHigh | Rated Interval validated | 93.4 | 2413 | 495 / 0 / 35 | 530 / 530 | 265 | — | 2026-09-30First try |
GPT-6 AstraE-9bf97597e76765ac | Medium | Rated Interval validated | 93.1 | 2395 | 493 / 0 / 37 | 530 / 530 | 265 | — | 2026-09-05First try |
Grok 4.7E-f78a8d93b3f3b22d | Medium | Rated Interval validated | 93.1 | 2395 | 493 / 0 / 37 | 530 / 530 | 265 | — | 2026-09-30First try |
Muse WonderE-3479ecef790ca818 | XHigh | Rated Interval validated | 93.1 | 2395 | 493 / 0 / 37 | 530 / 530 | 265 | — | 2026-10-12First try |
Claude Opus 5E-d7111b534ebcaba2 | Medium | Rated Interval validated | 92.6 | 2369 | 490 / 0 / 40 | 530 / 530 | 265 | — | 2026-07-25First try |
Qwen3.8 Max 0902E-31a5f2c0df745a1d | XHigh | Rated Interval validated | 92.4 | 2360 | 489 / 0 / 41 | 530 / 530 | 265 | — | 2026-10-09First try |
Claude Opus 5.5E-122583de77d058b0 | Low | Rated Interval validated | 91.9 | 2335 | 486 / 0 / 44 | 530 / 530 | 265 | — | 2026-09-22First try |
Pareto 26.10E-f2cb6b45243a0889 | Provider | Rated Interval validated | 91.9 | 2335 | 486 / 0 / 44 | 530 / 530 | 265 | — | 2026-10-21First try |
Claude Fable 5.1E-44ec1bd2d32e7f19 | Low | Rated Interval validated | 91.7 | 2327 | 485 / 0 / 45 | 530 / 530 | 265 | — | 2026-09-03First try |
GPT-6 SolE-65dcf2ee9a0cf52e | Medium | Rated Interval validated | 91.6 | 2319 | 484 / 0 / 46 | 530 / 530 | 265 | — | 2026-09-22First try |
GPT-6.1 SolE-e1b05e00d8a27c7b | Low | Rated Interval validated | 91.4 | 2311 | 483 / 0 / 47 | 530 / 530 | 265 | — | 2026-10-01First try |
Gemini 3.6 FlashE-8b9ff59d79afb558 | High | Rated Interval validated | 91.1 | 2296 | 481 / 0 / 49 | 530 / 530 | 265 | — | 2026-07-24First try |
Qwen3.8 MaxE-776235fa35035a89 | Minimal | Rated Interval validated | 90.6 | 2273 | 478 / 0 / 52 | 530 / 530 | 265 | — | 2026-08-08First try |
Gemini 4 ArgonE-8539952a4a6e0dd5 | Medium | Rated Interval validated | 90.4 | 2266 | 477 / 0 / 53 | 530 / 530 | 265 | — | 2026-10-21First try |