Benchmark menu

Per-game leaderboard

GB2-08: Tumbleweed

This page compares player programs for GB2-08: Tumbleweed in the intensive filter. Programs are ranked by their relative 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 player programs using DuelLab's encoded rules for Tumbleweed; they are not a broad measure of human play strength.

GB2-08: Tumbleweed by model and setting

Player programs are ranked by relative per-game score (0–100). Raw rating is shown as an advanced metric, alongside wins, draws, losses, and a per-game uncertainty index.

Reasoning level: Intensive Game: GB2-08: Tumbleweed
GB2-08: Tumbleweed - Intensive filter
Rank Model Score Contributing programs Raw Elo W / D / L Uncertainty
1Claude Fable 5100.012039.887 / 1 / 48.1
2Grok 4.591.311945.9148 / 2 / 200.0
3Claude Sonnet 583.611883.377 / 0 / 138.7
4GPT-5.6 Sol81.011857.875 / 0 / 178.1
5GPT-5.573.811788.858 / 12 / 228.1
6Claude Opus 572.321777.8119 / 0 / 3912.2
7Claude Opus 4.870.211756.363 / 0 / 239.9
8GPT-5.459.011649.452 / 11 / 239.9
9GPT-5.6 Terra51.511577.343 / 13 / 309.9
10Gemini 3.6 Flash50.921561.7115 / 97 / 860.0
11O345.331507.7130 / 187 / 1370.0
12Ox Alpha44.531570.525 / 10 / 785.3
13Hy3 Preview41.111476.231 / 26 / 377.5
14GPT 535.911425.215 / 59 / 285.5
15Inkling Small34.711468.60 / 16 / 174.1
16GPT-OSS 120B33.911409.93 / 61 / 1811.1
17Nemotron 3 Ultra 550B A55B33.821408.611 / 120 / 4110.0
18DeepSeek V4 Flash33.811406.81 / 69 / 228.1
19Qwen3.8 Max33.411453.42 / 12 / 471.0
20MiMo-V2.533.111405.62 / 65 / 315.6
21DeepSeek V4 Pro33.011403.02 / 55 / 1913.2
22Gemini 3.1 Flash Lite32.211391.34 / 66 / 228.1
23Step 3.7 Flash32.011392.70 / 63 / 1711.8
24Claude Opus 4.528.311343.017 / 93 / 680.0
25Inkling27.911339.12 / 101 / 670.0
26Gemini 3.5 Flash Lite26.811329.96 / 100 / 560.0
27GPT-5.6 Luna24.211316.02 / 51 / 359.2
28Gemini 3.5 Flash24.211316.00 / 57 / 299.9

Individual player programs

Each row is one accepted player program for GB2-08: Tumbleweed. A model can appear more than once when it produced more than one program. Score is the normalized 0–100 result; Elo and W / D / L show the underlying play record. Confidence and evidence can be blank when too few compatible matches are available. Draws at the move limit count; one-sided model faults are listed separately.

Showing 36 player programs.
Player IDModelStatusScoreEloConfidenceW / D / LRated / attemptedOpponentsCapped drawsFaults / playersGeneration
E-12891afa894223fdClaude Fable 5Rated100.0204087 / 1 / 492 / 924310 / 1Repaired · 2026-07-02
E-6a0c04b4335bfea0Claude Opus 4.5Waiting0 / 0 / 00 / 0000 / 1First try · 2026-07-02
E-877eec51affacb3aClaude Opus 4.5Rated28.3134317 / 93 / 68178 / 17861930 / 1First try · 2026-07-21
E-22dc6fe5ae5fb8b1Claude Opus 4.8Rated70.2175663 / 0 / 2386 / 864100 / 1First try · 2026-07-03
E-a40851c4f03704e5Claude Opus 5Rated83.1188265 / 0 / 1378 / 783300 / 1First try · 2026-07-25
E-f5914ca845403384Claude Opus 5Rated61.4167454 / 0 / 2680 / 803600 / 1First try · 2026-07-25
E-df0550ce48a0b895Claude Sonnet 5Rated83.6188377 / 0 / 1390 / 904200 / 1First try · 2026-08-09
E-a7310645c3b96434DeepSeek V4 FlashRated33.814071 / 69 / 2292 / 9246690 / 1Repaired · 2026-07-02
E-3a5b5c4507a989d3DeepSeek V4 ProRated33.014032 / 55 / 1976 / 7637550 / 1Repaired · 2026-07-02
E-b335c9f92035f80bGemini 3.1 Flash LiteRated32.213914 / 66 / 2292 / 9245660 / 1Repaired · 2026-07-02
E-fa332e578ae97122Gemini 3.5 FlashRated24.213160 / 57 / 2986 / 8643570 / 1Repaired · 2026-07-02
E-2a14d2f80ade1bdaGemini 3.5 Flash LiteRated26.813306 / 100 / 56162 / 162581000 / 2First try · 2026-07-24
E-620b5ca2bcd82419Gemini 3.6 FlashRated72.91771107 / 18 / 25150 / 15058180 / 1First try · 2026-07-24
E-efad9f3f4e9183a0Gemini 3.6 FlashRated29.013528 / 79 / 61148 / 14858790 / 1First try · 2026-07-24
E-ebca8ea756b4a4adGPT 5Rated35.9142515 / 59 / 28102 / 10251590 / 1First try · 2026-07-02
E-431ff31b9cd599e1GPT-5.4Rated59.0164952 / 11 / 2386 / 8642110 / 1First try · 2026-08-09
E-3e859a6b27767128GPT-5.5Rated73.8178958 / 12 / 2292 / 9242120 / 1First try · 2026-07-02
E-ca2978e1aca13fecGPT-5.6 LunaRated24.213162 / 51 / 3588 / 8842510 / 1First try · 2026-08-09
E-58adff67f80afed1GPT-5.6 SolRated81.0185875 / 0 / 1792 / 924200 / 1First try · 2026-08-09
E-b061663dc342c859GPT-5.6 TerraRated51.5157743 / 13 / 3086 / 8642130 / 1Repaired · 2026-07-10
E-2e3c01d0103fc439Nemotron 3 Ultra 550B A55BRated34.014098 / 62 / 2292 / 9245620 / 1Repaired · 2026-07-11
E-8c24e8bcbe9b1228Nemotron 3 Ultra 550B A55BRated33.714093 / 58 / 1980 / 8038580 / 1First try · 2026-07-02
E-2270b9df7d2b32ccO3Rated75.21794114 / 1 / 35150 / 1505810 / 1First try · 2026-07-03
E-9fbb8e6c69f575f4O3Rated25.513196 / 82 / 56144 / 14458820 / 1First try · 2026-07-03
E-afa6c09eebc859c4O3Rated35.2141010 / 104 / 46160 / 160601040 / 1First try · 2026-07-02
E-24a75273941e6ea3GPT-OSS 120BRated33.914103 / 61 / 1882 / 8239610 / 1Repaired · 2026-07-02
E-eef649e110a96b98Qwen3.8 MaxLimited data33.414532 / 12 / 418 / 188120 / 1First try · 2026-08-18
E-27b1ec2d9d5762a4Ox AlphaRated54.9167114 / 0 / 014 / 14700 / 1First try · 2026-08-22
E-ec8f506d56ff547eOx AlphaRated35.114813 / 7 / 414 / 14770 / 1First try · 2026-08-22
E-f84aa60bad19339aOx AlphaRated43.315608 / 3 / 314 / 14730 / 1First try · 2026-08-22
E-8b265019ada47f24Step 3.7 FlashRated32.013930 / 63 / 1780 / 8038630 / 1Repaired · 2026-07-02
E-81a5e5ef15f371dcHy3 PreviewRated41.1147631 / 26 / 3794 / 9446260 / 1First try · 2026-07-02
E-d21921560d198b28InklingRated27.913392 / 101 / 67170 / 170631010 / 1Repaired · 2026-07-24
E-888363fef76ee16aInkling SmallLimited data34.714690 / 16 / 117 / 178160 / 1Repaired · 2026-08-09
E-707dd709e358a173Grok 4.5Rated91.31946148 / 2 / 20170 / 1706420 / 1First try · 2026-07-24
E-5f664f9580459829MiMo-V2.5Rated33.114062 / 65 / 370 / 7035650 / 1Repaired · 2026-07-02