Benchmark menu

Per-game leaderboard

GB2-01: Go

This page shows the per-game leaderboard for GB2-01: Go in the xhigh variants. Entrants are ranked by their relative per-game score within this game.

GB2-01: Go

Go

Go is a large-grid territory-control game with stone placement, captures, ko restrictions, passing, and final area scoring.

Scores here compare generated programs on DuelLab's encoded rules for Go; they are not a broad measure of human play strength.

Entrants

Every admitted unique code identity for GB2-01: Go. Capped draws are rated; unilateral model faults are listed separately.

Showing 23 entrants.
EntrantModelStatusScoreRatingW / L / DRated / attemptedOpponentsCapped drawsModel faultsAdmitted pluginsRepairGenerated
E-8000131e469a8bb6Claude Fable 5Rated76.31896153 / 25 / 0178 / 17890001First pass2026-07-03
E-53c8add70ed7d2e7Claude Opus 4.5Rated30.0133619 / 90 / 67176 / 176886701First pass2026-07-02
E-3308d2952452916eClaude Opus 4.8Rated67.01783125 / 32 / 21178 / 178902101First pass2026-07-03
E-5e353d79b07fa294Claude Sonnet 5Rated70.31824127 / 26 / 25178 / 178902501First pass2026-07-03
E-b9882650a0cde9acDeepSeek V4 FlashRated32.3136440 / 97 / 39176 / 176883901First pass2026-07-02
E-cb8739e00cafec5bDeepSeek V4 FlashRated41.4147469 / 74 / 34177 / 177893401Repaired2026-07-02
E-e192048d4eaa3c43DeepSeek V4 ProRated49.0156676 / 71 / 29176 / 176882901Repaired2026-07-02
E-3e63b136f656e976GPT 5Rated39.8145576 / 70 / 30176 / 176883001First pass2026-07-02
E-5d8a6b14c27539abGPT-5.4Rated100.02182177 / 0 / 0177 / 17789001First pass2026-07-02
E-ca99b677101014a8GPT-5.4 MiniRated52.51608109 / 57 / 12178 / 178901201First pass2026-07-02
E-db1fc33e2a2467d5GPT-5.4 NanoRated39.1144633 / 70 / 73176 / 176887301First pass2026-07-09
E-fbb3f2dccb7c7fbaGPT-5.4 NanoRated12.711278 / 146 / 23177 / 177892301First pass2026-07-02
E-8c61292c31db2722GPT-5.5Rated69.21810126 / 9 / 40175 / 178884001First pass2026-07-02
E-38187ed58fb76057GPT-5.6 LunaRated76.81903157 / 19 / 0176 / 17688001First pass2026-07-10
E-9e683b3cd6c7f7c3GPT-5.6 SolRated78.71926158 / 16 / 0174 / 18087001First pass2026-07-10
E-36d36f4895d90b11GPT-5.6 TerraRated38.6144058 / 80 / 38176 / 176883801First pass2026-07-10
E-99c7b0d758665b66Mistral Medium 3.5Rated11.0110613 / 131 / 32176 / 176883201First pass2026-07-02
E-ce064a9927030b28Nemotron 3 Ultra 550B A55BRated30.0133633 / 105 / 42180 / 184924201Repaired2026-07-11
E-4b0ce2bc67b08d8aStep 3.7 FlashRated7.2106116 / 147 / 13176 / 176881301Repaired2026-07-02
E-29e8870e43d717e5Hy3 PreviewRated65.11761119 / 36 / 21176 / 176882101First pass2026-07-02
E-8478c64d84536b61MiMo-V2.5Rated20.612222 / 100 / 75177 / 177897501Repaired2026-07-02
E-5ab5088af70c19d6GLM-5.2Rated40.3146034 / 60 / 84178 / 180908401Repaired2026-07-11
E-88bed07a2e526eedGLM-5.2Rated47.9155331 / 25 / 120176 / 1768812001Repaired2026-07-03

GB2-01: Go leaderboard

Entrants are ranked by relative per-game score (0–100). Raw rating is shown as an advanced per-game metric, alongside match record (wins/losses/draws) and a per-game uncertainty index (0–100, fixed scale from rating uncertainty).

Reasoning level: XHigh Game: GB2-01: Go

Scroll sideways to see every column.

GB2-01: Go - XHigh variants
Rank Model Score Contributing programs Raw Elo W / L / D Uncertainty
1GPT-5.4100.012182.5177/0/00.0
2GPT-5.6 Sol78.711925.7158/16/00.0
3GPT-5.6 Luna76.811902.6157/19/00.0
4Claude Fable 576.311895.8153/25/00.0
5Claude Sonnet 570.311823.6127/26/250.0
6GPT-5.569.211809.8126/9/400.0
7Claude Opus 4.867.011782.8125/32/210.0
8Hy3 Preview65.111760.6119/36/210.0
9GPT-5.4 Mini52.511608.3109/57/120.0
10DeepSeek V4 Pro49.011566.576/71/290.0
11GLM-5.244.121506.565/85/2040.0
12GPT 539.811454.876/70/300.0
13GPT-5.6 Terra38.611439.758/80/380.0
14DeepSeek V4 Flash36.921419.0109/171/730.0
15Claude Opus 4.530.011336.219/90/670.0
16Nemotron 3 Ultra 550B A55B30.011335.833/105/420.0
17GPT-5.4 Nano25.921286.741/216/960.0
18MiMo-V2.520.611222.42/100/750.0
19Mistral Medium 3.511.011106.313/131/320.0
20Step 3.7 Flash7.211061.016/147/130.0