Qwen 3.7 Flash: benchmark results

Provider: Alibaba. Released 2026-07-25. Access: API.

Unified ELO 1635 ± 1, rank #165 of 1392 rated models, from 40 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MERA v2 - SAGE68.2Score (%)91.7
AI for Education Pedagogy - Science91.26Accuracy (%)85.5
MERA v2 - GorillaHard50.6Score (%)76.4
AI Chess Leaderboard (Reasoning)943Elo75.5
OTIS Mock AIME 2024-2586.67Accuracy (%)73.6
AI for Education Pedagogy - Primary88.73Accuracy (%)71.4
AI for Education Pedagogy - Secondary84.75Accuracy (%)71.4
AI for Education Pedagogy - Social studies83.64Accuracy (%)71.4
AI for Education Pedagogy - Maths84.92Accuracy (%)71.2
Chess Puzzles (Epoch AI)23Accuracy (%)69.6
AI for Education Pedagogy84.76Accuracy (%)68.7
MERA v20.41Total score63.9

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