Benchmark menu

Per-game leaderboard

GB2-05: Hex

This page shows the per-game leaderboard for GB2-05: Hex in the intensive filter. Entrants are ranked by their relative per-game score within this game.

GB2-05: Hex

Hex

Hex is a connection game on a hexagonal grid where players race to join opposite board edges, with a swap option after the first move.

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

Entrants

Every admitted unique code identity for GB2-05: Hex. 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 38 entrants.
EntrantModelStatusScoreEloConfidenceW / L / DRated / attemptedOpponentsCapped drawsFaults / pluginsGeneration
E-5aa4402ccce7b855Claude Fable 5Under-covered73.3156110 / 6 / 016 / 16800 / 1Repaired · 2026-07-02
E-ccacc93776fb7c88Claude Opus 4.5Under-covered26.815002 / 2 / 04 / 4200 / 1First pass · —
E-6774d700e60de31aClaude Opus 4.8Under-covered20.115312 / 0 / 02 / 2100 / 1Repaired · 2026-07-02
E-3f3686d02d1503faClaude Opus 5Under-covered100.0165712 / 0 / 012 / 12600 / 1First pass · 2026-07-25
E-a35dfc1adc62e4ebClaude Opus 5Under-covered41.014984 / 4 / 08 / 8400 / 1First pass · 2026-07-25
E-c902807aab849f31Claude Sonnet 5Under-covered37.915063 / 3 / 06 / 6300 / 1First pass · 2026-07-03
E-b306c8f5ffbbec38DeepSeek V4 FlashUnder-covered0.014690 / 2 / 02 / 2100 / 1Repaired · 2026-07-02
E-b3d889b0631db85fDeepSeek V4 FlashUnder-covered16.014661 / 3 / 04 / 4200 / 1Repaired · 2026-07-02
E-0ca4798825ceaccbDeepSeek V4 ProUnder-covered27.014552 / 6 / 08 / 8400 / 1Repaired · 2026-07-02
E-a72dd193e1dc4331Gemini 3.1 Flash LitePending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-02
E-b50da126631c1ea1Gemini 3.5 FlashUnder-covered19.015282 / 0 / 02 / 2100 / 1Repaired · 2026-07-02
E-27023858537764acGemini 3.5 Flash LitePending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-2aebd7296279d5e5Gemini 3.5 Flash LitePending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-68cd3c6f37a30f06Gemini 3.6 FlashPending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-c5ad048b6fb4e93dGPT 5Under-covered20.815332 / 0 / 02 / 2100 / 1First pass · 2026-07-02
E-f6403f7a30e19e82GPT-5.4Under-covered61.215477 / 3 / 010 / 10500 / 1First pass · 2026-07-02
E-481582e8f8e3217eGPT-5.4 MiniUnder-covered19.915312 / 0 / 02 / 2100 / 1First pass · 2026-07-02
E-95d1a872f51b78e7GPT-5.4 MiniUnder-covered10.314210 / 6 / 06 / 6300 / 1First pass · 2026-07-02
E-54b9ca4996a5589dGPT-5.4 NanoUnder-covered0.914720 / 2 / 02 / 2100 / 1First pass · 2026-07-02
E-b05a7e0987de9967GPT-5.4 NanoUnder-covered21.514252 / 8 / 010 / 10500 / 1First pass · 2026-07-09
E-762e9d80a60b41edGPT-5.5Under-covered52.815345 / 3 / 08 / 8400 / 1First pass · 2026-07-02
E-25a83ddaeb1268a4GPT-5.6 LunaUnder-covered24.414652 / 4 / 06 / 6300 / 1First pass · 2026-07-10
E-af9c2c2998294ad0GPT-5.6 SolUnder-covered7.814410 / 4 / 04 / 4200 / 1First pass · 2026-07-10
E-012ecb9ce0709c05GPT-5.6 TerraUnder-covered43.415234 / 2 / 06 / 6300 / 1Repaired · 2026-07-10
E-1954b83d439b455cMistral Medium 3.5Pending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-02
E-288f9a63e6b6cae3Kimi K3Under-covered36.415293 / 1 / 04 / 4200 / 1Repaired · 2026-07-18
E-95ac38f534c0080fKimi K3Under-covered26.014972 / 2 / 04 / 4200 / 1Repaired · 2026-07-17
E-48ed5c5e00ef38fbNemotron 3 Ultra 550B A55BPending0 / 0 / 00 / 0000 / 1First pass · 2026-07-02
E-da3b98638ff53ea9Nemotron 3 Ultra 550B A55BUnder-covered7.914420 / 4 / 04 / 4200 / 1Repaired · 2026-07-11
E-4fc62dc65156d576O3Pending0 / 0 / 00 / 0000 / 1First pass · 2026-07-03
E-358338a9fbdbcb6aStep 3.7 FlashPending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-02
E-92dc0ac61d798488Hy3 PreviewPending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-03
E-e9202aa1565d1c4eInklingPending0 / 0 / 00 / 0000 / 1Repaired · 2026-07-24
E-87d1c39ddc685d7bGrok 4.5Under-covered10.515021 / 1 / 02 / 2100 / 1Repaired · 2026-07-18
E-bd7fcc801178eba2Grok 4.5Under-covered0.014690 / 2 / 02 / 2100 / 1Repaired · 2026-07-18
E-235e4beedf22bc99GLM-5.2Under-covered45.915584 / 0 / 04 / 4200 / 1Repaired · 2026-07-11
E-503ecd021099d8f9GLM-5.2Under-covered0.014690 / 2 / 02 / 2100 / 1Repaired · 2026-07-02
E-51c8f79c1671819eGLM-5.2Under-covered25.514682 / 4 / 06 / 6300 / 1Repaired · 2026-07-02

GB2-05: Hex 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: GB2-05: Hex

Scroll sideways to see every column.

GB2-05: Hex - Intensive filter
Rank Model Score Contributing programs Raw Elo W / L / D Uncertainty
1Claude Opus 5100.011657.112/0/094.9
2Claude Fable 573.311561.010/6/077.5
3GPT-5.461.211547.37/3/0100.0
4GPT-5.552.811534.15/3/0100.0
5GLM-5.245.911558.44/0/0100.0
6GPT-5.6 Terra43.411523.24/2/0100.0
7Claude Opus 541.011497.84/4/0100.0
8Claude Sonnet 537.911506.33/3/0100.0
9Kimi K336.411529.23/1/0100.0
10DeepSeek V4 Pro27.011454.82/6/0100.0
11Claude Opus 4.526.811499.92/2/0100.0
12Kimi K326.011497.32/2/0100.0
13GLM-5.225.511468.12/4/0100.0
14GPT-5.6 Luna24.411464.62/4/0100.0
15GPT-5.4 Nano21.511425.22/8/0100.0
16GPT 520.811533.22/0/0100.0
17Claude Opus 4.820.111531.22/0/0100.0
18GPT-5.4 Mini19.911530.52/0/0100.0
19Gemini 3.5 Flash19.011527.92/0/0100.0
20DeepSeek V4 Flash16.011466.51/3/0100.0
21Grok 4.510.511501.61/1/0100.0
22GPT-5.4 Mini10.311421.40/6/0100.0
23Nemotron 3 Ultra 550B A55B7.911441.60/4/0100.0
24GPT-5.6 Sol7.811441.40/4/0100.0
25GPT-5.4 Nano0.911472.10/2/0100.0
26DeepSeek V4 Flash0.011469.50/2/0100.0
27GLM-5.20.011469.50/2/0100.0
28Grok 4.50.011469.30/2/0100.0