step-3.7-flash-20260528: benchmark results

Provider: StepFun. Access: API.

Aggregate Rating 1667 ± 40, rank #515 of 10916 rated models, from 121 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Evals for Every Language - Afrikaans (af)80.89Average Score (%)89.4
Evals for Every Language - Japanese (ja)74.96Average Score (%)85.1
Evals for Every Language - Kurdish (ku)70.15Average Score (%)84.4
Evals for Every Language - Bemba (bem)53.1Average Score (%)84.1
Evals for Every Language - Macedonian (mk)74.6Average Score (%)81.8
Evals for Every Language - Turkish (tr)75.97Average Score (%)81.5
Evals for Every Language - Yiddish (yi)72.57Average Score (%)80.3
Evals for Every Language - Sango (sg)45.79Average Score (%)77.8
Evals for Every Language - MMLU93.6Average Score (%)75
Evals for Every Language - Irish (ga)74.28Average Score (%)73.8
Evals for Every Language - Kazakh (kk)71.25Average Score (%)73.8
Evals for Every Language - Translation To22.67Average Score (%)72.5

Interactive version: theaggregate.ai/model?slug=step-3-7-flash-20260528 · How It Works · Data refreshed daily, snapshot 2026-11-08.