DuelLabGameBench 2
Menu

Games · GB2-01

Go

Go is a large-grid territory-control game with stone placement, captures, ko restrictions, passing, and final area scoring. Scores compare AI-written player programs on DuelLab's rules for Go; they do not measure human play strength.

Updated
2026-10-24
Reasoning settings
233
Player programs
259

Leaderboard

By model and setting

Each model and reasoning setting appears once. If a model produced more than one player program for this game, DuelLab averages them before ranking. 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.

GB2-01: Go - Reasoning Variants
Rank Model Reasoning Score Raw rating W / D / L
1GPT-5.4Medium97.52298.0499 / 7 / 10
2GPT-6 AstraXHigh97.32279.1501 / 0 / 15
3Gemini 4 ArgonHigh95.32163.6489 / 0 / 27
4GPT-5.4XHigh95.22159.7482 / 13 / 21
5Claude Opus 5.5Medium93.92103.6463 / 35 / 18
6Kimi K3.1 ProMax93.22079.1476 / 1 / 39
7Claude Fable 5.1XHigh93.22076.2475 / 2 / 39
8GPT-6 AstraLow93.22076.2464 / 24 / 28
9Gemini 4 ArgonLow92.52053.7472 / 0 / 44
10Muse Spark OpenXHigh92.32048.3471 / 0 / 45
11Pareto 26.10Provider91.42022.7466 / 0 / 50
12Claude Opus 5.5Low91.22015.4454 / 21 / 41
13Gemini 4 ArgonMedium91.12013.0464 / 0 / 52
14Gemini 3.5 FlashHigh90.51996.7450 / 21 / 45
15Claude Fable 5.1Low88.31943.9445 / 4 / 66
16Claude Fable 5XHigh86.21901.4436 / 1 / 79
17Muse Spark 1.2Medium85.81894.5423 / 23 / 70
18Muse WonderXHigh85.11881.0428 / 5 / 83
19Claude Opus 5.5XHigh84.41868.0422 / 9 / 85
20DeepSeek V4.1 Flash 1017High84.31866.4426 / 0 / 90
21Grok 4.5Medium83.91860.0402 / 44 / 70
22Grok 4.5High83.71856.9371 / 104 / 41
23GlimpseEnabled83.11846.1409 / 21 / 86
24GPT-5.6 LunaXHigh81.41819.6410 / 1 / 105
25GPT-5.5XHigh80.91812.5345 / 126 / 45

B1 ratings carry no uncertainty term, so no rating-evidence index is shown; see Methodology.

Underlying results

Individual player programs

Each row is one accepted player program for GB2-01: Go; 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.

259 player programs.

ModelReasoningStatusScoreRatingW / D / LMatchesOpponentsNotesGenerated
GPT-5.4E-d4fa7dbb9458c464MediumRated Interval validated97.52298499 / 7 / 10516 / 5162587 capped draws2026-08-09First try
GPT-6 AstraE-fc5055140bf78475XHighRated Interval validated97.32279501 / 0 / 15516 / 516258—2026-09-05First try
Gemini 4 ArgonE-c9b7bf650530d8baHighRated Interval validated95.32164489 / 0 / 27516 / 516258—2026-10-21First try
GPT-5.4E-9879a37f54e31cffXHighRated Interval validated95.22160482 / 13 / 21516 / 51625813 capped draws2026-08-09First try
Claude Opus 5.5E-9e6850818d33c592MediumRated Interval validated93.92104463 / 35 / 18516 / 51625835 capped draws2026-09-22First try
Gemini 3.6 FlashE-7484f4e7d8863a3bMediumRated Interval validated93.52088473 / 10 / 33516 / 51625810 capped draws2026-07-24First try
Kimi K3.1 ProE-030c793cb8126246MaxRated Interval validated93.22079476 / 1 / 39516 / 5162581 capped draw2026-10-24First try
Claude Fable 5.1E-443622d1ef9c6f19XHighRated Preliminary rev 976df3ba93.22076475 / 2 / 39516 / 5162582 capped draws2026-09-08First try
GPT-6 AstraE-7fbf151da7860da4LowRated Interval validated93.22076464 / 24 / 28516 / 51625824 capped draws2026-09-05First try
Gemini 4 ArgonE-b29fd1d6479f1bc8LowRated Interval validated92.52054472 / 0 / 44516 / 516258—2026-10-21First try
Muse Spark OpenE-482f21724d9e9898XHighRated Interval validated92.32048471 / 0 / 45516 / 516258—2026-10-20First try
Claude Opus 5E-700f3dddd611df07XHighRated Interval validated92.02038466 / 6 / 44516 / 5162586 capped draws2026-07-25First try
Pareto 26.10E-74fb208ee7158901ProviderRated Interval validated91.42023466 / 0 / 50516 / 516258—2026-10-21First try
Claude Opus 5.5E-c6c2699b24183df3LowRated Interval validated91.22015454 / 21 / 41516 / 51625821 capped draws2026-09-22First try
Gemini 4 ArgonE-31ef2459454f5e17MediumRated Interval validated91.12013464 / 0 / 52516 / 516258—2026-10-21First try
Gemini 3.5 FlashE-1802f43944b4b6f0HighRated Interval validated90.51997450 / 21 / 45516 / 51625821 capped draws2026-09-03First try
Claude Opus 5E-865d68ab219fd7eeNoneRated Interval validated90.41994428 / 64 / 24516 / 51625864 capped draws2026-07-25First try
Claude Fable 5.1E-d04c4d8635b641f9LowRated88.31944445 / 4 / 66515 / 5162584 capped draws2026-09-03First try
Claude Fable 5E-99c76fc062487714XHighRated Interval validated86.21901436 / 1 / 79516 / 5162581 capped draw2026-09-03First try
Muse Spark 1.2E-052cdf1e5975b860MediumRated Interval validated85.81895423 / 23 / 70516 / 51625823 capped draws2026-08-13First try
Muse WonderE-5c48ba07caee5554XHighRated Interval validated85.11881428 / 5 / 83516 / 5162585 capped draws2026-10-12First try
Claude Opus 5.5E-05f63acd714d31e5XHighRated Interval validated84.41868422 / 9 / 85516 / 5162589 capped draws2026-09-22First try
DeepSeek V4.1 Flash 1017E-5074fe2d98bd9ce2HighRated Interval validated84.31866426 / 0 / 90516 / 516258—2026-10-20First try
Grok 4.5E-d709aeb6802f9289MediumRated Interval validated83.91860402 / 44 / 70516 / 51625844 capped draws2026-07-24First try
Grok 4.5E-1c8e452ede41ea2eHighRated Interval validated83.71857371 / 104 / 41516 / 516258104 capped draws2026-07-24Repaired

Draws at the move cap count as draws; a crash or an illegal move counts as that program's loss and is also listed as a fault.