OWE-Bench - SmolLM2 Long Context: leaderboard

OWE-Bench - SmolLM2 Long Context — 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 March 2027. 14 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Fable 5.2Max0.7789#1
2Claude Sonnet 5.5Max0.6417#19
3Claude Opus 5.5Max0.53#5
4GPT-6 AstraMax0.505#6
5GPT-6.1 SolMax0.4511#10
6Gemini 4 CarbonHigh0.3672#2
7Claude Fable 5.1Max0.3528#15
8Gemini 4 ArgonHigh0.3345#12
9GLM-5.3Max0.2255#68
10DeepSeek V4.1 FlashMax0.1939#62
11Muse WonderMax0.02333#28
12Grok 4.7xHigh0.01889#47
13Beam 1.0Max0.01223#206
14InklingMax0#281

Interactive version: theaggregate.ai/benchmark?slug=owe-bench-smollm2-long-context · How It Works · Data refreshed daily, snapshot 2026-11-16.