Step-2-16k-Exp: benchmark results
Provider: StepFun. Access: API.
Aggregate Rating 1591 ± 28, rank #1104 of 10916 rated models, from 53 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Knowledge - Science - English (CompassBench 2501) | 91 | Score (%) | 100 |
| OpenCompass LLM - Instruction Following - English (CompassBench 2501) | 71.2 | Score (%) | 100 |
| OpenCompass LLM - Instruction Following (CompassBench 2501) | 67.3 | Score (%) | 97.4 |
| OpenCompass LLM - Language - English (CompassBench 2501) | 73.6 | Score (%) | 94.9 |
| OpenCompass Knowledge - Social Science - English (CompassBench 2501) | 91 | Score (%) | 93.6 |
| OpenCompass LLM - Knowledge - Chinese (CompassBench 2501) | 89.5 | Score (%) | 92.3 |
| OpenCompass LLM - Language (CompassBench 2501) | 71.1 | Score (%) | 92.3 |
| OpenCompass Language - Creation - English (CompassBench 2501) | 70.1 | Score (%) | 92.3 |
| OpenCompass Language - NLP - Chinese (CompassBench 2501) | 66.7 | Score (%) | 92.3 |
| OpenCompass Language - NLP - English (CompassBench 2501) | 80.8 | Score (%) | 92.3 |
| OpenCompass Reasoning - Academic - Chinese (CompassBench 2501) | 68.3 | Score (%) | 92.3 |
| OpenCompass Knowledge - Humanities - English (CompassBench 2501) | 87 | Score (%) | 91 |
Interactive version: theaggregate.ai/model?slug=step-2-16k-exp · How It Works · Data refreshed daily, snapshot 2026-11-08.