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.

GB2-07: Game of the Amazons - Reasoning Variants
Rank Model Reasoning Score Programs Raw rating Combined record (W / D / L)
1Claude Fable 5.2XHigh99.41 working of 1 attempt2927.5608 / 0 / 2
2Gemini 4 CarbonHigh97.71 working of 1 attempt2658.0596 / 0 / 14
3Claude Opus 5.5XHigh97.41 working of 1 attempt2629.7594 / 0 / 16
4GPT-6 AstraXHigh97.41 working of 1 attempt2629.7594 / 0 / 16
5GPT-6 AstraMedium96.51 working of 1 attempt2555.8588 / 0 / 22
6GPT-6.1 TerraXHigh96.51 working of 1 attempt2555.8588 / 0 / 22
7Gemini 4 HydrogenXHigh96.31 working of 1 attempt2544.7587 / 0 / 23
8Claude Fable 5.2Medium95.91 working of 1 attempt2512.9584 / 0 / 26
9Pareto 26.10 PreviewDefault95.61 working of 1 attempt2492.7582 / 0 / 28
10GPT-6.1 SolXHigh95.41 working of 1 attempt2482.9581 / 0 / 29
11GPT-6.1 Terra ProXHigh95.41 working of 1 attempt2482.9581 / 0 / 29
12GPT-6.1 TerraMedium95.31 working of 1 attempt2473.3580 / 0 / 30
13Pareto 26.10Default95.31 working of 1 attempt2473.3580 / 0 / 30
14Claude Opus 5XHigh94.61 working of 1 attempt2436.4576 / 0 / 34
15GPT-6.1 Terra ProNone93.91 working of 1 attempt2393.2571 / 0 / 39
16GPT-6.1 SolMedium93.71 working of 1 attempt2384.9570 / 0 / 40
17GPT-6.1 Terra ProMedium93.61 working of 1 attempt2376.6569 / 0 / 41
18Gemini 4 CarbonLow93.41 working of 1 attempt2368.5568 / 0 / 42
19GPT-6 AstraLow93.41 working of 1 attempt2368.5568 / 0 / 42
20Claude Fable 5.2Low93.31 working of 1 attempt2360.4567 / 0 / 43
21Claude Fable 5.1XHigh93.11 working of 1 attempt2352.4566 / 0 / 44
22GPT-6 SolMedium93.11 working of 1 attempt2352.4566 / 0 / 44
23GPT-6.1 SolLow93.01 working of 1 attempt2344.6565 / 0 / 45
24GPT-6 SolXHigh92.91 working of 1 attempt2336.7564 / 0 / 46
25Claude Fable 5.1Medium92.61 working of 1 attempt2321.4562 / 0 / 48

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.

305 player programs.

ModelReasoningStatusScoreRatingW / D / LMatchesOpponentsNotesGenerated
Claude Fable 5.2E-86e394c1f47b4cfbXHighRated Interval validated99.42927608 / 0 / 2610 / 610305—2026-11-09First try
Gemini 4 CarbonE-d972250f3a3dd214HighRated Interval validated97.72658596 / 0 / 14610 / 610305—2026-11-12First try
Claude Opus 5.5E-a2c3e4f63d5e1edbXHighRated Interval validated97.42630594 / 0 / 16610 / 610305—2026-09-22First try
GPT-6 AstraE-9ffb18636ed1a05fXHighRated Interval validated97.42630594 / 0 / 16610 / 610305—2026-09-05First try
GPT-6 AstraE-1c9e1c5642cb9d00MediumRated Interval validated96.52556588 / 0 / 22610 / 610305—2026-09-05First try
GPT-6.1 TerraE-b2e66ef6f350a55aXHighRated Interval validated96.52556588 / 0 / 22610 / 610305—2026-11-01First try
Gemini 4 HydrogenE-199717d1b80fddeaXHighRated Interval validated96.32545587 / 0 / 23610 / 610305—2026-11-12Repaired
Claude Fable 5.2E-23a35abdfac639beMediumRated Interval validated95.92513584 / 0 / 26610 / 610305—2026-11-09First try
Pareto 26.10 PreviewE-b0d14e6e532c3057DefaultRated Interval validated95.62493582 / 0 / 28610 / 610305—2026-10-09First try
GPT-6.1 SolE-19d39d632e696554XHighRated Interval validated95.42483581 / 0 / 29610 / 610305—2026-10-01First try
GPT-6.1 Terra ProE-3c86be35c6c17eb9XHighRated Interval validated95.42483581 / 0 / 29610 / 610305—2026-11-01First try
GPT-6.1 TerraE-1a13f3ccfc3da088MediumRated Interval validated95.32473580 / 0 / 30610 / 610305—2026-11-01First try
Pareto 26.10E-65eb86e302040a43DefaultRated Interval validated95.32473580 / 0 / 30610 / 610305—2026-10-21First try
Claude Opus 5E-1e882072d8d97548XHighRated Interval validated94.62436576 / 0 / 34610 / 610305—2026-07-25First try
GPT-6.1 Terra ProE-8a330cd9fe43f1dcNoneRated Interval validated93.92393571 / 0 / 39610 / 610305—2026-11-01First try
GPT-6.1 SolE-bc4cc89fd4bbead7MediumRated Interval validated93.72385570 / 0 / 40610 / 610305—2026-10-01First try
GPT-6.1 Terra ProE-47d9ee0b4c3fe93cMediumRated Interval validated93.62377569 / 0 / 41610 / 610305—2026-11-01First try
Gemini 4 CarbonE-5434c918630c00e0LowRated Interval validated93.42368568 / 0 / 42610 / 610305—2026-11-12First try
GPT-6 AstraE-b0b6ad65edfb8663LowRated Interval validated93.42368568 / 0 / 42610 / 610305—2026-09-05First try
Claude Fable 5.2E-33f782cd2be20120LowRated Interval validated93.32360567 / 0 / 43610 / 610305—2026-11-09First try
Claude Fable 5.1E-c7cbdce0356163ddXHighRated Interval validated93.12352566 / 0 / 44610 / 610305—2026-09-08First try
GPT-6 SolE-449cab4d6e898879MediumRated Interval validated93.12352566 / 0 / 44610 / 610305—2026-09-22First try
GPT-6.1 SolE-27b8b5b0c5b44a23LowRated Interval validated93.02345565 / 0 / 45610 / 610305—2026-10-01First try
GPT-6 SolE-048f1ff6d205f4c2XHighRated Interval validated92.92337564 / 0 / 46610 / 610305—2026-09-22First try
Claude Fable 5.1E-8669b1c5c68118e1MediumRated Interval validated92.62321562 / 0 / 48610 / 610305—2026-09-03First try