OWE-Bench - LVGL Renderer Speedup: leaderboard

OWE-Bench - LVGL Renderer Speedup — 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 August 2027. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Fable 5.2Max0.581#1
2Claude Opus 5.5Max0.5218#5
3Claude Sonnet 5.5Max0.4757#19
4GPT-6.1 SolMax0.4237#10
5Gemini 4 ArgonHigh0.4179#12
6GPT-6 AstraMax0.3936#6
7Claude Fable 5.1Max0.3723#15
8Gemini 4 CarbonHigh0.2698#2
9Muse WonderMax0.1979#28
10DeepSeek V4.1 FlashMax0.1746#62
11GLM-5.3Max0.167#68
12Grok 4.7xHigh0.08034#47
13Beam 1.0Max0.0434#206
14InklingMax0.002113#281

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