Qwen 3.5 9B Instruct: benchmark results
Provider: Alibaba. Released 2026-03-02.
Aggregate Rating 1642 ± 48, rank #675 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ProactBench - Emergent | 72.8 | Pass rate (%) on the Emergent triggers (turns 1-3: inferring | 93.3 |
| ProactBench - Weighted Score | 57 | Weighted score (%): Pass counts 1, Partial 0.5 and Fail 0, a | 73.3 |
| ProactBench - Recovery | 15.4 | Pass rate (%) on the Recovery triggers (turns 8-10: grounded | 66.7 |
| ProactBench - Critical | 51.4 | Pass rate (%) on the Critical triggers (turns 4-7: synthesis | 60 |
| Video-MME-v2 - Level 3 Complex Reasoning (w/o subs) | 10.04 | Level 3 Score w/o sub (%) | 41.7 |
| TAEF (Text Analytics Evaluation Framework) | 68.1 | F1 (self-reported) | 40 |
| Video-MME-v2 - Level 2 Temporal Understanding (w/o subs) | 9.87 | Level 2 Score w/o sub (%) | 22.9 |
| Video-MME-v2 | 26 | Avg Accuracy w/o sub (%) | 21.9 |
| Video-MME-v2 - Non-Lin Score (w/o subs) | 10.3 | Non-Lin Score w/o sub (%) | 15.6 |
| Video-MME-v2 - Level 1 Retrieval & Aggregation (w/o subs) | 11.46 | Level 1 Score w/o sub (%) | 2.1 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-9b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.