Qwen 3 Max: benchmark results

Alibaba's Qwen 3 Max, the proprietary API-only flagship of the Qwen 3 family (September 2025). Provider: Alibaba. Released 2025-09-24. Access: API.

Unified ELO 1631 ± 1, rank #179 of 1392 rated models, from 205 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Disentangling Language Roles in Multilingual L84JointSuccess (self-reported)100
SciEval - Astronomy75.64Astronomy (%)100
BacktestBench61.84Overall Accuracy (OA) (self-reported)95.5
SGI-Bench Wet Experiment33.62Wet Experiment Score94.4
LOM Benchmark - CycleDetection93Accuracy (%)91.7
WDCD R2 In-Document Resistance100Score (%)90
LLM Arena RU1113Arena Elo89.3
Every Act Has Its Price1004.4Care (self-reported)88.9
SGI-Bench31.97SGI-Score88.9
SciEval - Physics41.04Physics (%)88.9
BenchTable72.3Total Score (%)88.7
SnakeBench30.8TrueSkill Rating88.2

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