OWE-Bench - GPU4PySCF J/K Kernels: leaderboard

OWE-Bench - GPU4PySCF J/K Kernels — 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 2028. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Fable 5.2Max0.1671#1
2Claude Opus 5.5Max0.1444#5
3Gemini 4 CarbonHigh0.1289#2
4Claude Sonnet 5.5Max0.1035#19
5Claude Fable 5.1Max0.07598#15
6Gemini 4 ArgonHigh0.07443#12
7GPT-6.1 SolMax0.07433#10
8GPT-6 AstraMax0.06869#6
9Muse WonderMax0.007575#28
10DeepSeek V4.1 FlashMax0.005138#62
11GLM-5.3Max0.004013#68
12Grok 4.7xHigh0.00265#47
13Beam 1.0Max0.0001625#206
14InklingMax0#281

Interactive version: theaggregate.ai/benchmark?slug=owe-bench-gpu4pyscf-j-k-kernels · How It Works · Data refreshed daily, snapshot 2026-11-16.