OWE-Bench - Drawing to CAD: leaderboard

OWE-Bench - Drawing to CAD — 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 October 2027. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraMax0.4377#6
2Claude Fable 5.2Max0.4353#1
3GPT-6.1 SolMax0.425#10
4Claude Opus 5.5Max0.3732#5
5Gemini 4 CarbonHigh0.2961#2
6Claude Sonnet 5.5Max0.284#19
7Gemini 4 ArgonHigh0.1843#12
8Claude Fable 5.1Max0.1799#15
9Muse WonderMax0.07814#28
10DeepSeek V4.1 FlashMax0.0758#62
11Grok 4.7xHigh0.07015#47
12GLM-5.3Max0.05075#68
13Beam 1.0Max0.01956#206
14InklingMax0.002113#281

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