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 | XHigh | 99.6 | 1 working of 1 attempt | 3068.5 | 613 / 0 / 1 | >99% / 0% / <1% | |
| 2 | XHigh | 99.4 | 1 working of 1 attempt | 2981.7 | 611 / 0 / 3 | >99% / 0% / <1% | |
| 3 | XHigh | 99.1 | 1 working of 1 attempt | 2884.0 | 608 / 0 / 6 | 99% / 0% / 1% | |
| 4 | Low | 98.3 | 1 working of 1 attempt | 2730.5 | 601 / 0 / 13 | 98% / 0% / 2% | |
| 5 | XHigh | 97.9 | 1 working of 1 attempt | 2682.2 | 598 / 0 / 16 | 97% / 0% / 3% | |
| 6 | Medium | 96.4 | 1 working of 1 attempt | 2550.5 | 587 / 0 / 27 | 96% / 0% / 4% | |
| 7 | Medium | 96.4 | 2 working of 2 attempts | 2591.8 program range 2435–2748 | 1175 / 0 / 53 | 96% / 0% / 4% | |
| 8 | Low | 94.9 | 1 working of 1 attempt | 2464.5 | 577 / 0 / 37 | 94% / 0% / 6% | |
| 9 | High | 94.8 | 1 working of 1 attempt | 2457.0 | 576 / 0 / 38 | 94% / 0% / 6% | |
| 10 | Medium | 94.6 | 1 working of 1 attempt | 2449.6 | 575 / 0 / 39 | 94% / 0% / 6% | |
| 11 | Max | 93.5 | 1 working of 1 attempt | 2401.7 | 568 / 0 / 46 | 93% / 0% / 7% | |
| 12 | Medium | 93.4 | 1 working of 1 attempt | 2395.3 | 567 / 0 / 47 | 92% / 0% / 8% | |
| 13 | XHigh | 93.2 | 1 working of 1 attempt | 2389.0 | 566 / 0 / 48 | 92% / 0% / 8% | |
| 14 | XHigh | 92.9 | 1 working of 1 attempt | 2376.7 | 564 / 0 / 50 | 92% / 0% / 8% | |
| 15 | Default | 92.9 | 1 working of 1 attempt | 2376.7 | 564 / 0 / 50 | 92% / 0% / 8% | |
| 16 | Medium | 92.0 | 1 working of 1 attempt | 2341.8 | 558 / 0 / 56 | 91% / 0% / 9% | |
| 17 | XHigh | 91.4 | 1 working of 1 attempt | 2319.9 | 554 / 0 / 60 | 90% / 0% / 10% | |
| 18 | High | 90.9 | 1 working of 1 attempt | 2304.2 | 551 / 0 / 63 | 90% / 0% / 10% | |
| 19 | Medium | 90.5 | 1 working of 1 attempt | 2288.9 | 548 / 0 / 66 | 89% / 0% / 11% | |
| 20 | Default | 90.3 | 1 working of 1 attempt | 2283.9 | 547 / 0 / 67 | 89% / 0% / 11% | |
| 21 | XHigh | 89.9 | 1 working of 1 attempt | 2269.2 | 544 / 0 / 70 | 89% / 0% / 11% | |
| 22 | High | 89.7 | 1 working of 1 attempt | 2264.4 | 543 / 0 / 71 | 88% / 0% / 12% | |
| 23 | Max | 89.6 | 1 working of 1 attempt | 2259.6 | 542 / 0 / 72 | 88% / 0% / 12% | |
| 24 | Low | 89.4 | 1 working of 1 attempt | 2254.8 | 541 / 0 / 73 | 88% / 0% / 12% | |
| 25 | Medium | 89.3 | 1 working of 1 attempt | 2250.2 | 540 / 0 / 74 | 88% / 0% / 12% |
Underlying results
Individual player programs
Each row is one accepted player program for GB2-08: Tumbleweed; 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.
306 player programs.
| Model | Reasoning | Status | Score | Rating | W / D / L | Matches | Opponents | Notes | Generated |
|---|---|---|---|---|---|---|---|---|---|
Claude Fable 5.2E-eaf42da4142bf00a | XHigh | Rated Interval validated | 99.6 | 3069 | 613 / 0 / 1 | 614 / 614 | 307 | — | 2026-11-09First try |
Claude Sonnet 5.5E-541bf150c2f4631e | XHigh | Rated Interval validated | 99.4 | 2982 | 611 / 0 / 3 | 614 / 614 | 307 | — | 2026-09-30First try |
Gemini 4 HydrogenE-47f0b8c11cd9d787 | XHigh | Rated Interval validated | 99.1 | 2884 | 608 / 0 / 6 | 614 / 614 | 307 | — | 2026-11-12First try |
Claude Opus 5.5E-57a4de42441fcf4f | Medium | Rated Interval validated | 98.4 | 2748 | 602 / 0 / 12 | 614 / 614 | 307 | — | 2026-09-22First try |
Claude Fable 5.2E-97268de26e34bfb6 | Low | Rated Interval validated | 98.3 | 2730 | 601 / 0 / 13 | 614 / 614 | 307 | — | 2026-11-09First try |
Claude Fable 5.1E-3b200c9d604918bf | XHigh | Rated Interval validated | 97.9 | 2682 | 598 / 0 / 16 | 614 / 614 | 307 | — | 2026-09-08Repaired |
Claude Fable 5.2E-99cc8eca0260f823 | Medium | Rated Interval validated | 96.4 | 2551 | 587 / 0 / 27 | 614 / 614 | 307 | — | 2026-11-09First try |
Kimi K3.1 ProE-61e4f4af2980a72a | Low | Rated Interval validated | 94.9 | 2465 | 577 / 0 / 37 | 614 / 614 | 307 | — | 2026-10-24First try |
Gemini 4 CarbonE-f4088f09a1880156 | High | Rated Interval validated | 94.8 | 2457 | 576 / 0 / 38 | 614 / 614 | 307 | — | 2026-11-12First try |
GPT-6 AstraE-57bd25d1868a06af | Medium | Rated Interval validated | 94.6 | 2450 | 575 / 0 / 39 | 614 / 614 | 307 | — | 2026-09-05First try |
Claude Opus 5.5E-7bc967a8698a261b | Medium | Rated Interval validated | 94.3 | 2435 | 573 / 0 / 41 | 614 / 614 | 307 | — | 2026-09-22First try |
GLM-5.3E-934b7beff533b847 | Max | Rated Interval validated | 93.5 | 2402 | 568 / 0 / 46 | 614 / 614 | 307 | — | 2026-10-09First try |
Claude Fable 5E-6f759a260a58e8b0 | Medium | Rated Interval validated | 93.4 | 2395 | 567 / 0 / 47 | 614 / 614 | 307 | — | 2026-09-03First try |
GPT-6.1 SolE-ef4b18fe50fa33f5 | XHigh | Rated Interval validated | 93.2 | 2389 | 566 / 0 / 48 | 614 / 614 | 307 | — | 2026-10-01First try |
Claude Opus 5.5E-96d7941179f2fd25 | XHigh | Rated Interval validated | 92.9 | 2377 | 564 / 0 / 50 | 614 / 614 | 307 | — | 2026-09-22First try |
Pareto 26.10 PreviewE-a7625120083a55f5 | Default | Rated Interval validated | 92.9 | 2377 | 564 / 0 / 50 | 614 / 614 | 307 | — | 2026-10-09First try |
Claude Opus 5E-e73b6fb78880341d | Medium | Rated Interval validated | 92.5 | 2359 | 561 / 0 / 53 | 614 / 614 | 307 | — | 2026-07-25First try |
Gemini 4 CarbonE-0ef7205e8805b2d5 | Medium | Rated Interval validated | 92.0 | 2342 | 558 / 0 / 56 | 614 / 614 | 307 | — | 2026-11-12First try |
GPT-6 AstraE-02789044133a041d | XHigh | Rated Interval validated | 91.4 | 2320 | 554 / 0 / 60 | 614 / 614 | 307 | — | 2026-09-05First try |
Kimi K3.1 ProE-6e2f73a33800455b | High | Rated Interval validated | 90.9 | 2304 | 551 / 0 / 63 | 614 / 614 | 307 | — | 2026-10-24First try |
Grok 4.7E-0de6715ab5dcd98f | Medium | Rated Interval validated | 90.5 | 2289 | 548 / 0 / 66 | 614 / 614 | 307 | — | 2026-09-30First try |
Pareto 26.10E-9122d2a0e6db4097 | Default | Rated Interval validated | 90.3 | 2284 | 547 / 0 / 67 | 614 / 614 | 307 | — | 2026-10-21First try |
GPT-6.1 Terra ProE-436389744d5a9587 | XHigh | Rated Interval validated | 89.9 | 2269 | 544 / 0 / 70 | 614 / 614 | 307 | — | 2026-11-01First try |
GLM-5.3 FlashX LegalE-bab82bfd0a4a6098 | High | Rated Interval validated | 89.7 | 2264 | 543 / 0 / 71 | 614 / 614 | 307 | — | 2026-11-03First try |
Kimi K3.1 ProE-4627e664e705189a | Max | Rated Interval validated | 89.6 | 2260 | 542 / 0 / 72 | 614 / 614 | 307 | — | 2026-10-24First try |