Qwen 3.7 Max (Max) — benchmark results
Provider: Alibaba. Released 2026-05-20. Access: API.
Unified ELO 1907 ± 40, rank #60 of 1776 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HMMT February 2026 | 97.1 | Score (self-reported) | 100 |
| CritPt | 13.4 | Accuracy (self-reported) | 96.6 |
| AA-LCR | 69 | Score (self-reported) | 92.7 |
| ITBench-AA | 42.5 | Average Precision at Full Recall (self-reported) | 90.9 |
| MRCR-v2 128k | 90.4 | Score (self-reported) | 90.9 |
| MathArena - APEX 2025 | 44.5 | Accuracy (%) | 88.6 |
| MathArena - HMMT Feb 2025 | 95 | Accuracy (%) | 85.3 |
| IOI | 46.75 | Score (self-reported) | 83 |
| ResearchClawBench | 18.7 | Overall (self-reported) | 82.4 |
| IMO-AnswerBench | 90 | Score (self-reported) | 76.9 |
| MCP Atlas | 76.4 | Score (self-reported) | 71 |
| Vals Index | 57.49 | Score (self-reported) | 70.8 |
Interactive version: theaggregate.ai/model?slug=qwen-3-7-max-max · How the rankings work · Data refreshed daily, snapshot 2026-07-22.