Qwen 3.7 Max: benchmark results
Alibaba's flagship Qwen 3.7 model for broad high-end chat and reasoning workloads. Provider: Alibaba. Released 2026-05-20. Access: API.
Unified ELO 1729 ± 1, rank #26 of 1392 rated models, from 247 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HMMT February 2026 | 97.1 | Score (self-reported) | 100 |
| LLM Stats (HMMT Feb 26) | 97.1 | Score (%) | 100 |
| LLM Stats (MAXIFE) | 89.2 | Score (%) | 100 |
| LLM Stats (MMLU-ProX) | 87 | Score (%) | 100 |
| LLM Stats (PolyMATH) | 86.5 | Score (%) | 100 |
| Qwen3.7 Launch - Apex | 44.5 | Score (%) | 100 |
| Qwen3.7 Launch - GPQA Diamond | 92.4 | Score (%) | 100 |
| Qwen3.7 Launch - Global PIQA | 91.4 | Score (%) | 100 |
| Qwen3.7 Launch - IFBench | 79.1 | Score (%) | 100 |
| Qwen3.7 Launch - IMOAnswerBench | 90 | Score (%) | 100 |
| Qwen3.7 Launch - MRCR-v2 128k | 90.4 | Accuracy (%) | 100 |
| Qwen3.7 Launch - QwenSVG | 1608 | Elo | 100 |
Interactive version: theaggregate.ai/model?slug=qwen-3-7-max · How It Works · Data refreshed daily, snapshot 2026-09-05.