OWE-Bench - Design/Build/Fly Aircraft: leaderboard

OWE-Bench - Design/Build/Fly Aircraft — 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 February 2027. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraMax0.7153#6
2Claude Fable 5.2Max0.7143#1
3GPT-6.1 SolMax0.5706#10
4Claude Opus 5.5Max0.4247#5
5Gemini 4 CarbonHigh0.384#2
6Claude Sonnet 5.5Max0.3446#19
7Gemini 4 ArgonHigh0.2858#12
8Claude Fable 5.1Max0.2226#15
9Muse WonderMax0.09333#28
10GLM-5.3Max0.05563#68
11DeepSeek V4.1 FlashMax0.04131#62
12InklingMax0.01443#281
13Beam 1.0Max0.01018#206
14Grok 4.7xHigh0.0003625#47

Interactive version: theaggregate.ai/benchmark?slug=owe-bench-design-build-fly-aircraft · How It Works · Data refreshed daily, snapshot 2026-11-16.