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.

GB2-06: International Draughts - Reasoning Variants
Rank Model Reasoning Score Programs Raw rating Combined record (W / D / L)
1Claude Opus 5.5XHigh99.51 working of 1 attempt2886.0502 / 0 / 0
2Claude Fable 5.1XHigh98.21 working of 1 attempt2599.8492 / 4 / 6
3Gemini 4 ArgonHigh97.61 working of 1 attempt2526.6486 / 8 / 8
4Pareto 26.10Provider93.61 working of 1 attempt2292.7458 / 17 / 27
5Gemini 4 ArgonMedium92.71 working of 1 attempt2259.9449 / 25 / 28
6GLM-5.3High92.51 working of 1 attempt2250.7448 / 24 / 30
7Claude Opus 5.5Medium92.41 working of 1 attempt2247.6452 / 15 / 35
8Kimi K3.1 ProHigh92.11 working of 1 attempt2238.7448 / 20 / 34
9Kimi K3.1 ProLow92.11 working of 1 attempt2238.7451 / 14 / 37
10GPT-6.1 SolMedium91.91 working of 1 attempt2232.8445 / 24 / 33
11Claude Haiku 5.5XHigh91.81 working of 1 attempt2229.9437 / 39 / 26
12GLM-5.3Max91.11 working of 1 attempt2207.5440 / 25 / 37
13Gemini 4 ArgonLow90.41 working of 1 attempt2186.2441 / 15 / 46
14GPT-6 SolXHigh90.21 working of 1 attempt2181.0435 / 25 / 42
15Grok 4.7XHigh90.21 working of 1 attempt2181.0441 / 13 / 48
16GPT-6.1 SolXHigh89.91 working of 1 attempt2170.9427 / 37 / 38
17Claude Sonnet 5.5XHigh89.81 working of 1 attempt2168.4427 / 36 / 39
18Claude Opus 5None88.61 working of 1 attempt2136.9427 / 23 / 52
19Gemini 3.8 FlashHigh88.61 working of 1 attempt2136.9424 / 29 / 49
20GPT-6 AstraMedium88.61 working of 1 attempt2136.9427 / 23 / 52
21Claude Fable 5.1Medium88.41 working of 1 attempt2132.3430 / 15 / 57
22Claude Fable 5XHigh87.61 working of 1 attempt2111.8422 / 22 / 58
23Pareto 26.10 PreviewProvider87.51 working of 1 attempt2109.5413 / 39 / 50
24GPT-6 AstraXHigh87.01 working of 1 attempt2096.3415 / 29 / 58
25GLM-5.2High85.81 working of 1 attempt2068.5414 / 18 / 70

Individual player programs

Each row is one accepted player program for GB2-06: International Draughts; 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.

252 player programs.

ModelReasoningStatusScoreRatingW / D / LMatchesOpponentsNotesGenerated
Claude Opus 5.5E-53134a8de9454448XHighRated Interval validated99.52886502 / 0 / 0502 / 502251—2026-09-22First try
Claude Fable 5.1E-bba525f4f6adfed4XHighRated Interval validated98.22600492 / 4 / 6502 / 502251—2026-09-08First try
Gemini 4 ArgonE-ec915524f0712927HighRated Interval validated97.62527486 / 8 / 8502 / 502251—2026-10-21Repaired
Pareto 26.10E-5d01b5fd5968de69ProviderRated Interval validated93.62293458 / 17 / 27502 / 5022511 capped draw2026-10-21First try
Gemini 4 ArgonE-670f358c134cd627MediumRated Interval validated92.72260449 / 25 / 28502 / 5022517 capped draws2026-10-21First try
GLM-5.3E-256f22a4f85c8ce5HighRated Interval validated92.52251448 / 24 / 30502 / 5022513 capped draws2026-09-03First try
Claude Opus 5.5E-cb36bcb6ba10266bMediumRated Interval validated92.42248452 / 15 / 35502 / 5022511 capped draw2026-09-22First try
Kimi K3.1 ProE-02f59a1e4530ae4dHighRated Interval validated92.12239448 / 20 / 34502 / 5022516 capped draws2026-10-24First try
Kimi K3.1 ProE-65169852dc5bea55LowRated Interval validated92.12239451 / 14 / 37502 / 5022515 capped draws2026-10-24First try
GPT-6.1 SolE-a4495c1439ca4ab3MediumRated Interval validated91.92233445 / 24 / 33502 / 5022512 capped draws2026-10-01First try
Claude Haiku 5.5E-7508ffbcb30ce9d7XHighRated Interval validated91.82230437 / 39 / 26502 / 5022513 capped draws2026-10-09First try
Gemini 3.6 FlashE-905531e52a8460f5HighRated Interval validated91.32213449 / 9 / 44502 / 502251—2026-07-24Repaired
GLM-5.3E-206af7e7704349b5MaxRated Interval validated91.12207440 / 25 / 37502 / 502251—2026-10-09First try
Gemini 4 ArgonE-1f96b98702c94d80LowRated Interval validated90.42186441 / 15 / 46502 / 5022515 capped draws2026-10-21First try
GPT-6 SolE-81df9cc10626f8a6XHighRated Interval validated90.22181435 / 25 / 42502 / 502251—2026-09-22First try
Grok 4.7E-c0e64450a9e18936XHighRated Interval validated90.22181441 / 13 / 48502 / 502251—2026-09-30Repaired
GPT-6.1 SolE-767db8a17224b0adXHighRated Interval validated89.92171427 / 37 / 38502 / 5022512 capped draws2026-10-01First try
Claude Sonnet 5.5E-33fb3ee55eb8f02fXHighRated Interval validated89.82168427 / 36 / 39502 / 502251—2026-09-30First try
Claude Opus 5E-ca8985adf5006428NoneRated Interval validated88.62137427 / 23 / 52502 / 5022513 capped draws2026-07-25First try
Gemini 3.8 FlashE-174c84d35d15c4faHighRated Interval validated88.62137424 / 29 / 49502 / 5022515 capped draws2026-09-03First try
GPT-6 AstraE-7061a88204394bb0MediumRated Interval validated88.62137427 / 23 / 52502 / 5022514 capped draws2026-09-05First try
Claude Fable 5.1E-cca67b58479ba609MediumRated Interval validated88.42132430 / 15 / 57502 / 5022511 capped draw2026-09-03Repaired
Claude Fable 5E-df96158297e694e7XHighRated Interval validated87.62112422 / 22 / 58502 / 5022515 capped draws2026-09-03First try
Pareto 26.10 PreviewE-994a64010e669aecProviderRated Interval validated87.52110413 / 39 / 50502 / 5022511 capped draw2026-10-09First try
GPT-6 AstraE-abd8473bac4f3524XHighRated Interval validated87.02096415 / 29 / 58502 / 502251—2026-09-05First try