Qwen-VL-Max: benchmark results
Provider: Alibaba. Released 2024-01-25. Access: API.
Aggregate Rating 1591 ± 7, rank #475 of 1635 rated models, from 592 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| FinEval - Multimodal | 76.3 | Weighted average accuracy (%) | 100 |
| MMErroR - Error Type Classification - Biology & Healthcare | 66.39 | Error Type Classification Accuracy (%) | 100 |
| OpenVLM MME - Landmark | 191 | Score | 100 |
| OpenVLM MMBench V1.1 EN - Celebrity Recognition | 94.6 | Accuracy (%) | 96.3 |
| OpenVLM MME - Celebrity | 177.1 | Score | 95.7 |
| OpenVLM MMT-Bench - Counting By Visual Prompting | 50 | Score (%) | 95.4 |
| OpenVLM MMT-Bench - Religious Recognition | 80 | Score (%) | 94.9 |
| OpenVLM MMT-Bench - Logo and Brand Recognition | 100 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Social Relation Recognition | 70 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Rock Recognition | 75 | Score (%) | 94.2 |
| OpenVLM MME - Posters | 187.8 | Score | 94 |
| OpenVLM MMT-Bench - Animated Character Recognition | 70 | Score (%) | 93.7 |
Interactive version: theaggregate.ai/model?slug=qwen-vl-max · How It Works · Data refreshed daily, snapshot 2026-11-05.