Benchmark menu

Per-game leaderboard

GB2-03: Shogi

This page shows one pooled row per model and reasoning setting for GB2-03: Shogi. Repeated evaluated entrants are combined in the same way as the benchmark page's game filter.

GB2-03: Shogi

Shogi

Shogi is a 9x9 chess-family game with promotion zones, captured-piece drops, repetition handling, and impasse scoring.

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

GB2-03: Shogi leaderboard

Each model and reasoning setting appears once. Repeated evaluated entrants are pooled before ranking, matching the score shown when this game is selected on the benchmark page. Raw rating, match record, and uncertainty are pooled over those contributing entrants.

Reasoning level: Reasoning Variants Game: GB2-03: Shogi
GB2-03: Shogi - Reasoning Variants
Rank Model Reasoning Score Raw Elo W / L / D Uncertainty
1GPT-5.6 SolXHigh100.01647.919/4/156.2
2GPT-5.5XHigh87.41591.316/5/2132.5
3DeepSeek V4 FlashMedium67.21573.28/2/485.3
4GPT-5.5Medium57.51530.36/2/1168.1
5GPT 4.1None55.41534.76/2/781.2
6GPT 5Medium54.71492.110/12/2230.8
7GPT-5.4None53.01521.94/2/1174.1
8DeepSeek V4 FlashNone49.51514.211/7/1477.7
9Grok Build 0.1Medium47.51494.66/7/960.5
10GPT-5.4 MiniNone45.61525.19/4/1994.9
11Minimax M3None45.21519.93/2/6100.0
12DeepSeek V4 ProNone43.41493.26/6/574.1
13MiMo-V2.5-ProNone42.81475.74/7/1454.3
14GPT-5.6 TerraXHigh42.31546.74/1/1100.0
15Claude Opus 4.5None41.51504.32/2/894.9
16Hy3 PreviewNone41.51513.85/4/1100.0
17GPT-5.4 MiniMedium40.01514.78/4/1893.8
18GPT 5XHigh39.81495.23/4/689.9
19GLM-5.2Medium39.11519.32/1/5100.0
20North Mini CodeMedium37.71469.34/7/965.4
21Nex N2 ProMedium37.51475.64/8/574.1
22Gemini 3.5 FlashMedium37.21543.63/0/2100.0
23Nex N2 ProNone37.01465.21/4/1662.9
24GLM-5.2None36.91504.310/8/1589.4
25GPT-5.4Medium36.71473.13/5/974.1
26Claude Sonnet 5None36.41519.52/1/4100.0
27GPT-5.6 TerraMedium34.71456.51/5/1660.5
28Claude Opus 4.5Medium33.21547.53/0/1100.0
29GPT-5.5None30.21456.41/5/1077.5
30Nemotron 3 Ultra 550B A55BNone27.11501.41/1/4100.0
31Qwen3.7 MaxNone26.11488.71/2/4100.0
32GPT-5.6 SolMedium25.61545.83/0/0100.0
33MiMo-V2.5None24.11520.21/0/3100.0
34Gemini 3.5 FlashXHigh22.91516.61/0/3100.0
35Claude Opus 4.8Medium22.61515.82/1/1100.0
36GPT-5.6 LunaNone21.91476.12/4/1100.0
37Claude Sonnet 5XHigh21.51444.53/7/294.9
38GPT-5.4XHigh21.21435.94/8/285.3
39GPT-5.6 LunaXHigh19.21505.71/1/2100.0
40GPT-5.6 TerraNone16.81482.80/1/4100.0
41GPT-5.4 NanoNone14.81431.46/25/1377.0
42GLM-5.2XHigh12.61458.10/3/3100.0
43GPT-5.6 SolNone12.51506.61/1/1100.0
44GPT-5.4 NanoMedium12.01456.30/3/3100.0
45GPT-5.4 NanoXHigh10.31387.55/13/265.4
46GPT-OSS 120BMedium0.51470.90/2/1100.0
47Claude Fable 5XHigh0.51501.70/0/2100.0
48GPT-5.6 LunaMedium0.01469.40/2/1100.0

Entrants

Every admitted unique code identity for GB2-03: Shogi. Capped draws are rated; unilateral model faults are listed separately.

Showing 57 entrants.
EntrantModelReasoningStatusScoreRatingW / L / DRated / attemptedOpponentsCapped drawsModel faultsAdmitted pluginsRepairGenerated
E-04fa5b762ca703f0claude-fable-5highestunder covered0.515020 / 0 / 22 / 82061first pass2026-07-03T17:50:00Z
E-1975db22db58e9b0claude-opus-4-5noneunder covered41.515042 / 2 / 812 / 2312341first pass2026-07-02T16:51:45Z
E-cef4890e88806969claude-opus-4-5mediumunder covered33.215483 / 0 / 14 / 224091first pass2026-07-02T16:51:45Z
E-823b961281482cd1claude-opus-4-8mediumunder covered22.615162 / 1 / 14 / 144151first pass2026-07-02T16:51:45Z
E-2e43b0d163a7f12dclaude-sonnet-5highestunder covered21.514453 / 7 / 212 / 2312051first pass2026-07-03T16:50:00Z
E-e7d1e6857fc48e7aclaude-sonnet-5noneunder covered36.415192 / 1 / 47 / 247351first pass2026-07-02T16:51:45Z
E-0f669ae82234cf76cohere/north-mini-codemediumunder covered37.714694 / 7 / 920 / 3916041repaired2026-07-02T16:51:45Z
E-32e04fe5ad231857deepseek/deepseek-v4-flashmediumunder covered67.215738 / 2 / 414 / 2312131first pass2026-07-01T00:15:48Z
E-4e52c100ed276347deepseek/deepseek-v4-flashnoneunder covered39.014865 / 6 / 415 / 2114131first pass2026-07-02T06:10:40Z
E-ae9d59ec3a8a1d63deepseek/deepseek-v4-flashnoneunder covered60.015436 / 1 / 1017 / 2717041repaired2026-07-02T16:51:45Z
E-f82a334a0fd1fb8edeepseek/deepseek-v4-prononeunder covered43.414936 / 6 / 517 / 2717031repaired2026-07-02T16:51:45Z
E-46c3bb79de67d970google/gemini-3.5-flashhighestunder covered22.915171 / 0 / 34 / 1440101repaired2026-07-02T16:51:45Z
E-8ed795ea8fd1e0aagoogle/gemini-3.5-flashmediumunder covered37.215443 / 0 / 25 / 1950111repaired2026-07-02T16:51:45Z
E-c7da9914f86ee846gpt-4.1noneunder covered55.415356 / 2 / 715 / 2515131repaired2026-07-02T16:51:45Z
E-cc568b7069d5a0fdgpt-5highestunder covered39.814953 / 4 / 613 / 1713031first pass2026-07-02T16:51:45Z
E-e724ba162431d43agpt-5mediumrated54.7149210 / 12 / 2244 / 4731001first pass2026-07-02T16:51:45Z
E-496f83ad0bd69447gpt-5.4mediumunder covered36.714733 / 5 / 917 / 2017031first pass2026-07-02T16:51:45Z
E-4b75acaf58f14113gpt-5.4highestunder covered21.214364 / 8 / 214 / 2912131first pass2026-07-02T16:51:45Z
E-ebd456dc6b320c61gpt-5.4noneunder covered53.015224 / 2 / 1117 / 2917191first pass2026-07-02T16:51:45Z
E-35117a2e624d7970gpt-5.4-mininoneunder covered41.015030 / 0 / 1212 / 22120101first pass2026-07-02T16:51:45Z
E-3e164d6fc6c16cbbgpt-5.4-minimediumunder covered31.215142 / 1 / 36 / 136251first pass2026-07-01T00:15:48Z
E-7cbcc752e71e2d3bgpt-5.4-minimediumunder covered29.714852 / 3 / 49 / 199441first pass2026-07-02T06:10:40Z
E-9e39bdd48b7328c7gpt-5.4-minimediumunder covered59.115464 / 0 / 1115 / 2315371repaired2026-07-02T16:51:45Z
E-a51cd55f211305c2gpt-5.4-mininoneunder covered57.915496 / 2 / 513 / 1613031repaired2026-07-01T00:15:48Z
E-dfa5098ec9c3cf47gpt-5.4-mininoneunder covered37.815243 / 2 / 27 / 107031first pass2026-07-02T06:10:40Z
E-0ad5302188ed9e2fgpt-5.4-nanomediumunder covered12.014560 / 3 / 36 / 96031first pass2026-07-01T00:15:48Z
E-0ecb5b73c09f13b2gpt-5.4-nanononeunder covered10.213894 / 14 / 119 / 2514021first pass2026-07-01T00:15:48Z
E-6c4158c55a35d268gpt-5.4-nanohighestunder covered10.313885 / 13 / 220 / 2514011repaired2026-07-02T06:10:40Z
E-e4eac938bdd5d40egpt-5.4-nanononeunder covered25.214301 / 8 / 1221 / 3215111first pass2026-07-02T06:10:40Z
E-f2a7b43bf689e68egpt-5.4-nanononeunder covered9.014751 / 3 / 04 / 94031first pass2026-07-09T17:30:50Z
E-446036823932df16gpt-5.5noneunder covered30.214561 / 5 / 1016 / 2516181first pass2026-07-02T16:51:45Z
E-c002b0ea38b845f8gpt-5.5mediumunder covered57.515306 / 2 / 1119 / 2819161first pass2026-07-02T16:51:45Z
E-e53a932f820d9724gpt-5.5highestrated87.4159116 / 5 / 2142 / 5231001first pass2026-07-02T16:51:45Z
E-0300c20cb034e3a4gpt-5.6-lunamediumunder covered0.014690 / 2 / 13 / 103041first pass2026-07-10T05:30:29Z
E-3bca3b4e02f2acf9gpt-5.6-lunahighestunder covered19.215061 / 1 / 24 / 94031first pass2026-07-10T10:07:13Z
E-d192bb239087b778gpt-5.6-lunanoneunder covered21.914762 / 4 / 17 / 167051first pass2026-07-10T05:30:29Z
E-2ac4e83a1b9ba989gpt-5.6-solhighestunder covered100.0164819 / 4 / 124 / 3118011first pass2026-07-10T05:30:29Z
E-527999604d620080gpt-5.6-solnoneunder covered12.515071 / 1 / 13 / 93041first pass2026-07-10T05:30:29Z
E-b741de451c385e8egpt-5.6-solmediumunder covered25.615463 / 0 / 03 / 93031first pass2026-07-10T05:30:29Z
E-39e763b7d49209b4gpt-5.6-terrahighestunder covered42.315474 / 1 / 16 / 106031repaired2026-07-10T05:30:29Z
E-c2f0f8f062659e83gpt-5.6-terranoneunder covered16.814830 / 1 / 45 / 125131first pass2026-07-10T10:07:13Z
E-e44c62bb88a3042agpt-5.6-terramediumunder covered34.714571 / 5 / 1622 / 3417131first pass2026-07-10T05:30:29Z
E-44cc758f104a74abminimax/minimax-m3noneunder covered45.215203 / 2 / 611 / 2111041first pass2026-07-02T16:51:45Z
E-57ad2863cbf32d58nex-agi/nex-n2-promediumunder covered37.514764 / 8 / 517 / 2417131repaired2026-07-02T16:51:45Z
E-a05d4cef5f788f16nex-agi/nex-n2-prononeunder covered37.014651 / 4 / 1621 / 3618441repaired2026-07-02T16:51:45Z
E-eb6e8a61f620acfdnvidia/nemotron-3-ultra-550b-a55bnoneunder covered27.115011 / 1 / 46 / 186091repaired2026-07-02T16:51:45Z
E-87b4bd61da369b42openai/gpt-oss-120bmediumunder covered0.514710 / 2 / 13 / 173091repaired2026-07-02T16:51:45Z
E-3325ebad2ef000c1qwen/qwen3.7-maxnoneunder covered26.114891 / 2 / 47 / 127041first pass2026-07-02T16:51:45Z
E-2346cd049910cc0atencent/hy3-previewnoneunder covered41.515145 / 4 / 110 / 2610031first pass2026-07-02T16:51:45Z
E-2a93343a8a88fd59x-ai/grok-build-0.1mediumunder covered47.514956 / 7 / 922 / 3419031repaired2026-07-02T16:51:45Z
E-b10e12eb6aef112exiaomi/mimo-v2.5noneunder covered24.115201 / 0 / 34 / 154191first pass2026-07-02T16:51:45Z
E-b0948c64dddf57adxiaomi/mimo-v2.5-prononerated42.814764 / 7 / 1425 / 3822431repaired2026-07-02T16:51:45Z
E-40aa28533b1afda1z-ai/glm-5.2mediumunder covered39.115192 / 1 / 58 / 128231repaired2026-07-11T17:22:09Z
E-4cab1ccd88bbeb2ez-ai/glm-5.2noneunder covered58.015325 / 2 / 1219 / 2319031first pass2026-07-02T16:51:45Z
E-583be34ca1fb2631z-ai/glm-5.2noneunder covered21.814862 / 3 / 16 / 126061repaired2026-07-11T14:00:45Z
E-bd2653f21938737fz-ai/glm-5.2highestunder covered12.614580 / 3 / 36 / 116141repaired2026-07-11T15:43:48Z
E-f1af7b938699d971z-ai/glm-5.2noneunder covered31.014953 / 3 / 28 / 208181repaired2026-07-02T06:10:40Z