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 |  Claude Opus 5.5 | XHigh | 97.8 | 1 working of 1 attempt | 2659.0 | 522 / 0 / 8 |
| 2 |  GPT-6 Astra | XHigh | 97.8 | 1 working of 1 attempt | 2659.0 | 522 / 0 / 8 |
| 3 | GPT-6 Astra | Medium | 96.7 | 1 working of 1 attempt | 2552.9 | 516 / 0 / 14 |
| 4 | Claude Opus 5 | XHigh | 95.9 | 1 working of 1 attempt | 2495.8 | 512 / 0 / 18 |
| 5 |  Pareto 26.10 Preview | Provider | 95.9 | 1 working of 1 attempt | 2495.8 | 512 / 0 / 18 |
| 6 | GPT-6.1 Sol | XHigh | 95.9 | 1 working of 1 attempt | 2495.8 | 512 / 0 / 18 |
| 7 | GPT-6.1 Sol | Medium | 95.7 | 1 working of 1 attempt | 2482.6 | 511 / 0 / 19 |
| 8 | Pareto 26.10 | Provider | 95.1 | 1 working of 1 attempt | 2445.2 | 508 / 0 / 22 |
| 9 | GPT-6 Astra | Low | 94.6 | 1 working of 1 attempt | 2410.4 | 505 / 0 / 25 |
| 10 | GPT-6.1 Sol | Low | 94.6 | 1 working of 1 attempt | 2410.4 | 505 / 0 / 25 |
| 11 | GPT-6 Sol | XHigh | 94.4 | 1 working of 1 attempt | 2399.3 | 504 / 0 / 26 |
| 12 | GPT-6 Sol | Medium | 94.0 | 1 working of 1 attempt | 2377.8 | 502 / 0 / 28 |
| 13 | Claude Fable 5.1 | XHigh | 93.9 | 2 working of 2 attempts | 2367.5 program range 2357–2378 | 1002 / 0 / 58 |
| 14 | Claude Fable 5.1 | Medium | 93.9 | 1 working of 1 attempt | 2367.4 | 501 / 0 / 29 |
| 15 | Claude Sonnet 5.5 | XHigh | 93.7 | 1 working of 1 attempt | 2357.1 | 500 / 0 / 30 |
| 16 |  Kimi K3 | Max | 93.7 | 1 working of 1 attempt | 2357.1 | 500 / 0 / 30 |
| 17 | Claude Opus 5.5 | Medium | 93.5 | 1 working of 1 attempt | 2347.0 | 499 / 0 / 31 |
| 18 | Claude Opus 5.5 | Low | 91.9 | 1 working of 1 attempt | 2263.9 | 490 / 0 / 40 |
| 19 |  MiMo-V2.6-Pro | Enabled | 89.8 | 1 working of 1 attempt | 2170.8 | 478 / 0 / 52 |
| 20 | GPT-5.6 Sol | XHigh | 88.7 | 1 working of 1 attempt | 2130.6 | 472 / 0 / 58 |
| 21 | Claude Fable 5.1 | Low | 88.2 | 1 working of 1 attempt | 2111.9 | 469 / 0 / 61 |
| 22 |  Muse Wonder | Medium | 87.5 | 1 working of 1 attempt | 2088.2 | 465 / 0 / 65 |
| 23 | GPT-5.5 | XHigh | 87.0 | 1 working of 1 attempt | 2071.3 | 462 / 0 / 68 |
| 24 | Kimi K3.1 Pro | High | 85.1 | 1 working of 1 attempt | 2014.9 | 451 / 0 / 79 |
| 25 | Claude Opus 5 | Medium | 84.3 | 2 working of 2 attempts | 2048.9 program range 1825–2273 | 892 / 0 / 168 |
Underlying results
Individual player programs
Each row is one accepted player program for GB2-07: Game of the Amazons; 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 5.5E-a2c3e4f63d5e1edb | XHigh | Rated Interval validated | 97.8 | 2659 | 522 / 0 / 8 | 530 / 530 | 265 | — | 2026-09-22First try |
GPT-6 AstraE-9ffb18636ed1a05f | XHigh | Rated Interval validated | 97.8 | 2659 | 522 / 0 / 8 | 530 / 530 | 265 | — | 2026-09-05First try |
GPT-6 AstraE-1c9e1c5642cb9d00 | Medium | Rated Interval validated | 96.7 | 2553 | 516 / 0 / 14 | 530 / 530 | 265 | — | 2026-09-05First try |
Claude Opus 5E-1e882072d8d97548 | XHigh | Rated Interval validated | 95.9 | 2496 | 512 / 0 / 18 | 530 / 530 | 265 | — | 2026-07-25First try |
Pareto 26.10 PreviewE-b0d14e6e532c3057 | Provider | Rated Interval validated | 95.9 | 2496 | 512 / 0 / 18 | 530 / 530 | 265 | — | 2026-10-09First try |
GPT-6.1 SolE-19d39d632e696554 | XHigh | Rated Interval validated | 95.9 | 2496 | 512 / 0 / 18 | 530 / 530 | 265 | — | 2026-10-01First try |
GPT-6.1 SolE-bc4cc89fd4bbead7 | Medium | Rated Interval validated | 95.7 | 2483 | 511 / 0 / 19 | 530 / 530 | 265 | — | 2026-10-01First try |
Pareto 26.10E-65eb86e302040a43 | Provider | Rated Interval validated | 95.1 | 2445 | 508 / 0 / 22 | 530 / 530 | 265 | — | 2026-10-21First try |
GPT-6 AstraE-b0b6ad65edfb8663 | Low | Rated Interval validated | 94.6 | 2410 | 505 / 0 / 25 | 530 / 530 | 265 | — | 2026-09-05First try |
GPT-6.1 SolE-27b8b5b0c5b44a23 | Low | Rated Interval validated | 94.6 | 2410 | 505 / 0 / 25 | 530 / 530 | 265 | — | 2026-10-01First try |
GPT-6 SolE-048f1ff6d205f4c2 | XHigh | Rated Interval validated | 94.4 | 2399 | 504 / 0 / 26 | 530 / 530 | 265 | — | 2026-09-22First try |
Claude Fable 5.1E-c7cbdce0356163dd | XHigh | Rated Interval validated | 94.0 | 2378 | 502 / 0 / 28 | 530 / 530 | 265 | — | 2026-09-08First try |
GPT-6 SolE-449cab4d6e898879 | Medium | Rated Interval validated | 94.0 | 2378 | 502 / 0 / 28 | 530 / 530 | 265 | — | 2026-09-22First try |
Claude Fable 5.1E-8669b1c5c68118e1 | Medium | Rated Interval validated | 93.9 | 2367 | 501 / 0 / 29 | 530 / 530 | 265 | — | 2026-09-03First try |
Claude Fable 5.1E-11d32603b1221b2e | XHigh | Rated Interval validated | 93.7 | 2357 | 500 / 0 / 30 | 530 / 530 | 265 | — | 2026-09-03First try |
Claude Sonnet 5.5E-295743189005f049 | XHigh | Rated Interval validated | 93.7 | 2357 | 500 / 0 / 30 | 530 / 530 | 265 | — | 2026-09-30First try |
Kimi K3E-dadbcf081f688390 | Max | Rated Interval validated | 93.7 | 2357 | 500 / 0 / 30 | 530 / 530 | 265 | — | 2026-10-09First try |
Claude Opus 5.5E-393b10028aaa43a6 | Medium | Rated Interval validated | 93.5 | 2347 | 499 / 0 / 31 | 530 / 530 | 265 | — | 2026-09-22First try |
Claude Opus 5E-54bca3ce32fa85ae | Medium | Rated Interval validated | 92.1 | 2273 | 491 / 0 / 39 | 530 / 530 | 265 | — | 2026-07-25First try |
Claude Opus 5.5E-2771a403cc3d9cab | Low | Rated Interval validated | 91.9 | 2264 | 490 / 0 / 40 | 530 / 530 | 265 | — | 2026-09-22First try |
MiMo-V2.6-ProE-25837f164a15a28a | Enabled | Rated Interval validated | 89.8 | 2171 | 478 / 0 / 52 | 530 / 530 | 265 | — | 2026-10-09First try |
GPT-5.6 SolE-475950cead6d8909 | XHigh | Rated Interval validated | 88.7 | 2131 | 472 / 0 / 58 | 530 / 530 | 265 | — | 2026-08-09First try |
Claude Fable 5.1E-8d5af37f6619d723 | Low | Rated Interval validated | 88.2 | 2112 | 469 / 0 / 61 | 530 / 530 | 265 | — | 2026-09-03First try |
Muse WonderE-dfdbd0d95d49d1e8 | Medium | Rated Interval validated | 87.5 | 2088 | 465 / 0 / 65 | 530 / 530 | 265 | — | 2026-10-12First try |
GPT-5.5E-6201278165db862a | XHigh | Rated Interval validated | 87.0 | 2071 | 462 / 0 / 68 | 530 / 530 | 265 | — | 2026-07-02First try |