Qwen 3.7 Plus: benchmark results

Alibaba's balanced Qwen 3.7 Plus tier. Provider: Alibaba. Released 2026-06-02. Access: API.

Unified ELO 1688 ± 1, rank #72 of 1392 rated models, from 191 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Stats (DeepPlanning)62.3Score (%)100
LLM Stats (MLVU)87.4Score (%)100
LLM Stats (OCRBench_V2)67.1Score (%)100
LLM Stats (SimpleVQA)81.7Score (%)100
LLM Stats (WorldVQA)61.1Score (%)100
OCRBench-V2 (en)70.7Score (self-reported)100
OCRBench-V2 (zh)67.1Score (self-reported)100
AA IFBench77.96Accuracy (%)96.8
VideoMME w sub.88Score (self-reported)96.4
LLM Stats (PolyMATH)84Score (%)96.2
RealWorldQA86.9RealWorldQA (self-reported)95.8
LLM Stats (MMLU-ProX)85.4Score (%)94.3

Interactive version: theaggregate.ai/model?slug=qwen-3-7-plus · How It Works · Data refreshed daily, snapshot 2026-09-05.