OWE-Bench - Storm Surge Forecast: leaderboard

OWE-Bench - Storm Surge Forecast — Expert-built long-horizon science, software, ML and engineering tasks with task-specific hardware and4–18hour budgets; native baseline/ceiling normalized held-out reward,mean of8attempts and equal task means. Web frozen October1,2026;deterministic sealed verifier. Native model/effort and every attempt retained. Score fraction0–1;higher is better.

Metric: Mean normalized held-out task score (fraction, 0–1). Source: openworldengineer.mercor.com. Saturation forecast: Around November 2027. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraMax0.6119#6
2Claude Fable 5.2Max0.5044#1
3GPT-6.1 SolMax0.4781#10
4Gemini 4 CarbonHigh0.4659#2
5Claude Opus 5.5Max0.4561#5
6Gemini 4 ArgonHigh0.37#12
7Claude Sonnet 5.5Max0.3406#19
8Claude Fable 5.1Max0.3383#15
9Muse WonderMax0.318#28
10Grok 4.7xHigh0.1957#47
11DeepSeek V4.1 FlashMax0.1778#62
12GLM-5.3Max0.1541#68
13Beam 1.0Max0.08959#206
14InklingMax0.04634#281

Interactive version: theaggregate.ai/benchmark?slug=owe-bench-storm-surge-forecast · How It Works · Data refreshed daily, snapshot 2026-11-16.