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
2026-10-16
Reasoning variants
209
Player programs
235

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-5.4Medium98.02363.7674 / 8 / 10
2GPT-6 AstraXHigh97.22287.6599 / 0 / 19
3GPT-5.4XHigh95.32178.2646 / 17 / 29
4Claude Opus 5.5Medium93.42099.0422 / 30 / 16
5GPT-6 AstraLow93.02086.0554 / 28 / 36
6Claude Fable 5.1XHigh92.92083.8566 / 3 / 49
7Claude Opus 5.5Low90.82021.3414 / 21 / 33
8Gemini 3.5 FlashHigh90.62015.6535 / 28 / 53
9Claude Fable 5.1Low89.11979.3538 / 3 / 76
10Claude Fable 5XHigh87.41940.3526 / 0 / 90
11Muse WonderXHigh85.71907.3397 / 5 / 66
12Muse Spark 1.2Medium85.01894.7553 / 30 / 105
13Grok 4.5High85.01894.3541 / 152 / 57
14Claude Opus 5.5XHigh84.81890.3391 / 8 / 69
15Grok 4.5Medium84.41883.4455 / 48 / 71
16GPT-5.4None82.11845.4516 / 65 / 115
17GPT-5.6 LunaXHigh82.01843.8544 / 1 / 145
18Muse Spark 1.2XHigh81.61838.6379 / 2 / 87
19Grok 4.7Medium80.11815.0357 / 31 / 80
20GPT-5.6 SolNone79.51805.8528 / 2 / 164
21Claude Sonnet 5.5XHigh79.21802.9297 / 143 / 28
22Hy3 PreviewNone79.11800.8402 / 95 / 83
23GPT-5.5XHigh78.91797.7356 / 152 / 54
24GPT-5.6 TerraMedium78.41790.4461 / 5 / 154
25Ember-1Max78.11786.8359 / 8 / 101

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.

235 player programs.

ModelReasoningStatusScoreRatingW / D / LMatchesOpponentsNotesGenerated
GPT-5.4E-d4fa7dbb9458c464MediumRated Interval validated98.02364674 / 8 / 10692 / 6922348 capped draws2026-08-09First try
GPT-6 AstraE-fc5055140bf78475XHighRated Interval validated97.22288599 / 0 / 19618 / 618234—2026-09-05First try
GPT-5.4E-9879a37f54e31cffXHighRated Interval validated95.32178646 / 17 / 29692 / 69223417 capped draws2026-08-09First try
Gemini 3.6 FlashE-7484f4e7d8863a3bMediumRated Interval validated93.52104526 / 9 / 35570 / 5702349 capped draws2026-07-24First try
Claude Opus 5.5E-9e6850818d33c592MediumRated Interval validated93.42099422 / 30 / 16468 / 46823430 capped draws2026-09-22First try
GPT-6 AstraE-7fbf151da7860da4LowRated Interval validated93.02086554 / 28 / 36618 / 61823428 capped draws2026-09-05First try
Claude Fable 5.1E-443622d1ef9c6f19XHighRated Preliminary rev 976df3ba92.92084566 / 3 / 49618 / 6182343 capped draws2026-09-08First try
Claude Opus 5.5E-c6c2699b24183df3LowRated Interval validated90.82021414 / 21 / 33468 / 46823421 capped draws2026-09-22First try
Gemini 3.5 FlashE-1802f43944b4b6f0HighRated Interval validated90.62016535 / 28 / 53616 / 61623428 capped draws2026-09-03First try
Claude Opus 5E-700f3dddd611df07XHighRated Interval validated90.22006485 / 6 / 55546 / 5462346 capped draws2026-07-25First try
Claude Opus 5E-865d68ab219fd7eeNoneRated Interval validated89.21980475 / 74 / 31580 / 58023474 capped draws2026-07-25First try
Claude Fable 5.1E-d04c4d8635b641f9LowRated89.11979538 / 3 / 76617 / 6182343 capped draws2026-09-03First try
Claude Fable 5E-99c76fc062487714XHighRated Interval validated87.41940526 / 0 / 90616 / 616234—2026-09-03First try
Muse WonderE-5c48ba07caee5554XHighRated Interval validated85.71907397 / 5 / 66468 / 4682345 capped draws2026-10-12First try
Muse Spark 1.2E-052cdf1e5975b860MediumRated Interval validated85.01895553 / 30 / 105688 / 68823430 capped draws2026-08-13First try
Grok 4.5E-1c8e452ede41ea2eHighRated Interval validated85.01894541 / 152 / 57750 / 750234152 capped draws2026-07-24Repaired
Claude Opus 5.5E-05f63acd714d31e5XHighRated Interval validated84.81890391 / 8 / 69468 / 4682348 capped draws2026-09-22First try
Grok 4.5E-d709aeb6802f9289MediumRated Interval validated84.41883455 / 48 / 71574 / 57423448 capped draws2026-07-24First try
Claude Opus 5E-93a6875407c22137MediumRated Interval validated82.11846433 / 38 / 91562 / 56223438 capped draws2026-07-25First try
GPT-5.4E-ed2ff19b1858f2c4NoneRated Interval validated82.11845516 / 65 / 115696 / 69623465 capped draws2026-08-09First try
GPT-5.6 LunaE-65def22b992d7636XHighRated Interval validated82.01844544 / 1 / 145690 / 6902341 capped draw2026-08-09First try
Muse Spark 1.2E-0f49665fd79b2646XHighRated Interval validated81.61839379 / 2 / 87468 / 4682342 capped draws2026-10-09First try
Grok 4.7E-4e41c6896cd841aeMediumRated Interval validated80.11815357 / 31 / 80468 / 46823431 capped draws2026-09-30First try
GPT-5.6 SolE-72b9641e74736185NoneRated Interval validated79.51806528 / 2 / 164694 / 6942342 capped draws2026-08-09First try
Claude Sonnet 5.5E-125c49d283c26a40XHighRated Interval validated79.21803297 / 143 / 28468 / 468234143 capped draws2026-09-30First try

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