Qwen2-VL Instruct (7B): benchmark results
Provider: Alibaba. Access: Open.
Aggregate Rating 1574 ± 23, rank #1336 of 10916 rated models, from 136 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM VHELM - Blink Reasoning | 82.89 | PEM | 100 |
| HELM VHELM - Mmmu - Materials | 63.33 | Prefix Quasi-Exact Match (%) | 98.9 |
| HELM VHELM - Unicorn | 90.15 | PEM | 97.9 |
| HELM VHELM - Unicorn - OODCV-VQA | 83.1 | Prefix Quasi-Exact Match (%) | 97.9 |
| HELM VHELM - Unicorn - Sketchy-VQA | 97.2 | Prefix Quasi-Exact Match (%) | 97.9 |
| HELM VHELM - Blink Reasoning - IQ Test | 50.67 | Prefix Quasi-Exact Match (%) | 93.6 |
| HELM VHELM - Mmmu - Mechanical Engineering | 63.33 | Prefix Quasi-Exact Match (%) | 93.6 |
| HELM VHELM - Pope | 89.1 | PEM | 93.6 |
| HELM VHELM - Vqa Base | 86.3 | PEM | 93.6 |
| HELM VHELM - Blink Reasoning - Multi-view Reasoning | 100 | Prefix Quasi-Exact Match (%) | 90.4 |
| HELM VHELM - Viz Wiz | 78.9 | PEM | 89.4 |
| HELM VHELM - Blink Perception - Relative Depth | 90.32 | Prefix Quasi-Exact Match (%) | 88.3 |
Interactive version: theaggregate.ai/model?slug=qwen2-vl-instruct-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.