Qwen3.5-397B-A17B-Think: benchmark results

Provider: Alibaba.

Aggregate Rating 1751 ± 43, rank #175 of 10916 rated models, from 5 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Video-MME-v248.8Avg Accuracy w/o sub (%)91.7
Video-MME-v2 - Level 2 Temporal Understanding (w/o subs)36.5Level 2 Score w/o sub (%)91.7
Video-MME-v2 - Level 3 Complex Reasoning (w/o subs)22.93Level 3 Score w/o sub (%)91.7
Video-MME-v2 - Non-Lin Score (w/o subs)30.3Non-Lin Score w/o sub (%)91.7
Video-MME-v2 - Level 1 Retrieval & Aggregation (w/o subs)35.43Level 1 Score w/o sub (%)89.6

Interactive version: theaggregate.ai/model?slug=qwen3-5-397b-a17b-think · How It Works · Data refreshed daily, snapshot 2026-11-08.