benchgap
Alibaba (Qwen) · model

Qwen3.7 Flash benchmark scores

As of 2026-10-29, Qwen3.7 Flash (Alibaba (Qwen)) has measured scores on 1 benchmark and estimated scores on 5 more.

BenchmarkScoreSource
AIME2692.1%estimated ± 1.4 pp, medium confidence
FrontierMath v2 (Tier 4)0.0%estimated ± 4.3 pp, medium confidence
FrontierMath v2 (Tiers 1-3)19.3%measured
HMMT Feb 202682.3%estimated ± 3.3 pp, medium confidence
HMMT Nov 202588.8%estimated ± 1.5 pp, low confidence
MMAnswerBench78.7%estimated ± 2.4 pp, low confidence