Benchmark menu

Per-game leaderboard

GB2-08: Tumbleweed

This page shows the per-game leaderboard for GB2-08: Tumbleweed in the xhigh variants. Entrants are ranked by their relative per-game score within this game.

GB2-08: Tumbleweed

Tumbleweed

Tumbleweed is a hex-grid territory game where players grow visible stacks and final control is determined by line-of-sight strength.

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

Entrants

Every admitted unique code identity for GB2-08: Tumbleweed. Capped draws are rated; unilateral model faults are listed separately.

Showing 29 entrants.
EntrantModelStatusScoreRatingW / L / DRated / attemptedOpponentsCapped drawsModel faultsAdmitted pluginsRepairGenerated
E-12891afa894223fdClaude Fable 5Rated98.22132184 / 4 / 1189 / 18995101Repaired2026-07-02
E-6a0c04b4335bfea0Claude Opus 4.5Rated71.31883162 / 27 / 0189 / 18995001First pass2026-07-02
E-22dc6fe5ae5fb8b1Claude Opus 4.8Rated87.02029176 / 13 / 0189 / 18995001First pass2026-07-03
E-b5805acb90227d50Claude Sonnet 5Rated24.4144833 / 39 / 114186 / 1869311401First pass2026-07-03
E-a7310645c3b96434DeepSeek V4 FlashRated16.113708 / 49 / 129186 / 1869312901Repaired2026-07-02
E-3a5b5c4507a989d3DeepSeek V4 ProRated15.0136111 / 48 / 127186 / 1869312701Repaired2026-07-02
E-b335c9f92035f80bGemini 3.1 Flash LiteRated16.713766 / 53 / 127186 / 1869312701Repaired2026-07-02
E-fa332e578ae97122Gemini 3.5 FlashRated5.512720 / 81 / 105186 / 1869310501Repaired2026-07-02
E-ebca8ea756b4a4adGPT 5Rated19.5140222 / 55 / 109186 / 1879310901First pass2026-07-02
E-ddc1f3e68bf1e74aGPT-5.4Rated50.21687114 / 39 / 35188 / 188953501First pass2026-07-02
E-1e5d07fc67fd8710GPT-5.4 MiniRated54.01722133 / 40 / 17190 / 190961701Repaired2026-07-02
E-b63acb1096519102GPT-5.4 MiniRated22.7143222 / 42 / 123187 / 1879412301First pass2026-07-02
E-98e55f26cfe52488GPT-5.4 NanoRated100.02149185 / 4 / 0189 / 18995001First pass2026-07-02
E-ba371b627387f273GPT-5.4 NanoRated16.513742 / 47 / 138187 / 1879413801First pass2026-07-02
E-bb4437fab8180bfcGPT-5.4 NanoRated21.1141718 / 48 / 120186 / 1869312001Repaired2026-07-09
E-3e859a6b27767128GPT-5.5Rated56.71747113 / 12 / 64189 / 189956401First pass2026-07-02
E-0fb7de4e8156a229GPT-5.6 LunaRated62.61802150 / 29 / 9188 / 18894901First pass2026-07-10
E-3ae62f195160add3GPT-5.6 SolRated3.8125640 / 123 / 25188 / 188942501First pass2026-07-10
E-b061663dc342c859GPT-5.6 TerraRated43.61626120 / 36 / 32188 / 188943201Repaired2026-07-10
E-2e3c01d0103fc439Nemotron 3 Ultra 550B A55BRated16.8137711 / 49 / 129189 / 1929612901Repaired2026-07-11
E-8c24e8bcbe9b1228Nemotron 3 Ultra 550B A55BRated18.2139012 / 50 / 127189 / 1899612701First pass2026-07-02
E-1b6fc7dfefb8cea4O3Rated75.21919162 / 19 / 8189 / 18995801First pass2026-07-03
E-24a75273941e6ea3GPT-OSS 120BRated16.6137510 / 46 / 131187 / 1879413101Repaired2026-07-02
E-8b265019ada47f24Step 3.7 FlashRated20.014077 / 43 / 136186 / 1869313601Repaired2026-07-02
E-81a5e5ef15f371dcHy3 PreviewRated23.7144159 / 73 / 58190 / 190965801First pass2026-07-02
E-5f664f9580459829MiMo-V2.5Rated14.513551 / 52 / 133186 / 1869313301Repaired2026-07-02
E-907d60010bf80864GLM-5.2Rated84.52005174 / 15 / 0189 / 19296001Repaired2026-07-11
E-9e8f3e43bfd2810eGLM-5.2Rated13.913500 / 64 / 122186 / 1869312201Repaired2026-07-02
E-bc28f817090c9f60GLM-5.2Rated0.012211 / 115 / 73189 / 189957301Repaired2026-07-02

GB2-08: Tumbleweed 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: GB2-08: Tumbleweed

Scroll sideways to see every column.

GB2-08: Tumbleweed - XHigh variants
Rank Model Score Contributing programs Raw Elo W / L / D Uncertainty
1Claude Fable 598.212132.2184/4/10.0
2Claude Opus 4.887.012028.7176/13/00.0
3O375.211919.3162/19/80.0
4Claude Opus 4.571.311882.6162/27/00.0
5GPT-5.6 Luna62.611802.5150/29/90.0
6GPT-5.556.711747.3113/12/640.0
7GPT-5.450.211687.0114/39/350.0
8GPT-5.4 Nano45.931646.9205/99/2580.0
9GPT-5.6 Terra43.611626.0120/36/320.0
10GPT-5.4 Mini38.321576.8155/82/1400.0
11GLM-5.232.831525.5175/194/1950.0
12Claude Sonnet 524.411447.933/39/1140.0
13Hy3 Preview23.711441.259/73/580.0
14Step 3.7 Flash20.011407.17/43/1360.0
15GPT 519.511402.022/55/1090.0
16Nemotron 3 Ultra 550B A55B17.521383.723/99/2560.0
17Gemini 3.1 Flash Lite16.711376.16/53/1270.0
18GPT-OSS 120B16.611375.210/46/1310.0
19DeepSeek V4 Flash16.111370.58/49/1290.0
20DeepSeek V4 Pro15.011360.711/48/1270.0
21MiMo-V2.514.511355.41/52/1330.0
22Gemini 3.5 Flash5.511272.40/81/1050.0
23GPT-5.6 Sol3.811256.340/123/250.0