Benchmark menu

Per-game leaderboard

GB2-02: Chess

This page shows the per-game leaderboard for GB2-02: Chess in the intensive filter. Entrants are ranked by their relative per-game score within this game.

GB2-02: Chess

Chess

Chess is an 8x8 tactical piece-movement game with asymmetric pieces, castling, en passant, promotion, check, checkmate, and automatic draw rules.

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

Entrants

Every admitted unique code identity for GB2-02: Chess. Elo and W/L/D use the public historical sequential-Elo result universe. Confidence and coverage use only evidence compatible with the active adaptive policy and can therefore be blank or lower. Capped draws are rated; unilateral model faults are listed separately.

Showing 32 entrants.
EntrantModelStatusScoreEloConfidenceW / L / DRated / attemptedOpponentsCapped drawsFaults / pluginsGeneration
E-f6a30e7d9986c9b8Claude Fable 5Under-covered69.215133 / 2 / 712 / 12600 / 1First pass · 2026-07-03
E-63fde27ceab11930Claude Opus 4.5Under-covered37.914850 / 1 / 56 / 6300 / 1First pass · —
E-a81bd03e764818f6Claude Opus 4.8Under-covered100.015738 / 2 / 414 / 14700 / 1First pass · 2026-07-03
E-6b45cf5aebf0c66eClaude Opus 5Under-covered40.714512 / 6 / 412 / 12600 / 1First pass · 2026-07-25
E-fcdd98d74e80134aClaude Opus 5Under-covered77.815325 / 4 / 312 / 12600 / 1First pass · 2026-07-25
E-e35d82f3fd09a1f4Claude Sonnet 5Under-covered64.215033 / 3 / 612 / 12600 / 1First pass · 2026-07-03
E-12dcb25370550495DeepSeek V4 FlashUnder-covered38.314680 / 2 / 68 / 8400 / 1Repaired · 2026-07-02
E-f2e6dd87c5656ee3DeepSeek V4 FlashUnder-covered13.715131 / 0 / 12 / 2100 / 1First pass · 2026-07-02
E-939a2dc80becc700Gemini 3.5 Flash LitePending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-a3da6c33bd873fb7Gemini 3.5 Flash LitePending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-5bccb9b671ce5eaeGPT 5Under-covered14.615151 / 0 / 12 / 2100 / 1First pass · 2026-07-02
E-4aa582caf9d9f6daGPT-5.4Under-covered69.215135 / 3 / 412 / 12600 / 1First pass · 2026-07-02
E-59f362f57892ab54GPT-5.4 MiniUnder-covered36.414820 / 1 / 56 / 6300 / 1First pass · 2026-07-02
E-7ef67fa468b6c12cGPT-5.4 MiniUnder-covered35.714810 / 1 / 56 / 6300 / 1First pass · 2026-07-02
E-1ce420518fc32de1GPT-5.4 NanoUnder-covered44.915010 / 0 / 66 / 6300 / 1Repaired · 2026-07-09
E-b084ba147f956b49GPT-5.4 NanoUnder-covered44.715001 / 1 / 46 / 6300 / 1Repaired · 2026-07-02
E-a30e455ba172d80eGPT-5.6 LunaUnder-covered7.915000 / 0 / 22 / 2100 / 1Repaired · 2026-07-10
E-33dc9365cfacf9c0GPT-5.6 SolUnder-covered99.415718 / 3 / 314 / 14700 / 1First pass · 2026-07-10
E-1b12482867c807c9GPT-5.6 TerraUnder-covered44.214512 / 6 / 614 / 14700 / 1First pass · 2026-07-10
E-1beeecccf3f2c84fKimi K3Pending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-97d6a124a1509fc1Kimi K3Under-covered45.815031 / 1 / 46 / 6300 / 1Repaired · 2026-07-18
E-b0de9fa15f139d69Kimi K3Under-covered38.015131 / 0 / 34 / 4200 / 1Repaired · 2026-07-17
E-2404d10bab26a254Nemotron 3 Ultra 550B A55BUnder-covered8.515010 / 0 / 22 / 2100 / 1Repaired · 2026-07-02
E-abd2cae3e3ff9bdbNemotron 3 Ultra 550B A55BUnder-covered7.915000 / 0 / 22 / 2100 / 1Repaired · 2026-07-11
E-e5b4a3586049e1c9GPT-OSS 120BUnder-covered0.014830 / 1 / 12 / 2100 / 1Repaired · 2026-07-02
E-fbf6af15f65f01d7Hy3 PreviewUnder-covered7.314990 / 0 / 22 / 2100 / 1Repaired · 2026-07-02
E-9468e854340a4e9bInklingPending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-0576181f89efa7a7Grok 4.5Under-covered15.315161 / 0 / 12 / 2100 / 1Repaired · 2026-07-24
E-8728b59f87c3369eGrok 4.5Under-covered1.314850 / 1 / 12 / 2100 / 1Repaired · 2026-07-17
E-4442c0e8a06187bfGLM-5.2Under-covered42.214540 / 4 / 812 / 12600 / 1Repaired · 2026-07-11
E-98394109d97d67eeGLM-5.2Under-covered24.914850 / 1 / 34 / 4200 / 1Repaired · 2026-07-02
E-dd5764d86cd88052GLM-5.2Under-covered56.615082 / 1 / 58 / 8400 / 1First pass · 2026-07-02

GB2-02: Chess 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: Intensive Game: GB2-02: Chess

Scroll sideways to see every column.

GB2-02: Chess - Intensive filter
Rank Model Score Contributing programs Raw Elo W / L / D Uncertainty
1Claude Opus 4.8100.011572.78/2/485.3
2GPT-5.6 Sol99.411571.38/3/385.3
3Claude Opus 577.811532.15/4/394.9
4Claude Fable 569.211513.33/2/794.9
5GPT-5.469.211513.35/3/494.9
6Claude Sonnet 564.211502.53/3/694.9
7GLM-5.256.611508.32/1/5100.0
8Kimi K345.811502.71/1/4100.0
9GPT-5.4 Nano44.911500.70/0/6100.0
10GPT-5.4 Nano44.711500.21/1/4100.0
11GPT-5.6 Terra44.211451.32/6/685.3
12GLM-5.242.211454.50/4/894.9
13Claude Opus 540.711451.32/6/494.9
14DeepSeek V4 Flash38.311468.50/2/6100.0
15Kimi K338.011513.31/0/3100.0
16Claude Opus 4.537.911485.40/1/5100.0
17GPT-5.4 Mini36.411482.20/1/5100.0
18GPT-5.4 Mini35.711480.60/1/5100.0
19GLM-5.224.911484.80/1/3100.0
20Grok 4.515.311516.01/0/1100.0
21GPT 514.611514.51/0/1100.0
22DeepSeek V4 Flash13.711512.51/0/1100.0
23Nemotron 3 Ultra 550B A55B8.511501.30/0/2100.0
24Nemotron 3 Ultra 550B A55B7.911500.00/0/2100.0
25GPT-5.6 Luna7.911500.00/0/2100.0
26Hy3 Preview7.311498.60/0/2100.0
27Grok 4.51.311485.50/1/1100.0
28GPT-OSS 120B0.011482.70/1/1100.0