QTuneVL1-2B: benchmark results
Provider: Other.
Aggregate Rating 1409 ± 29, rank #5669 of 10916 rated models, from 148 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMBench V1.1 CN - Image Scene | 91 | Accuracy (%) | 91.2 |
| OpenVLM MMBench V1.1 EN - Social Relation | 90.1 | Accuracy (%) | 90.3 |
| OpenVLM OCRBench - Handwritten Mathematical Expression Recognition | 77 | Correct Answers (%) | 89.3 |
| OpenVLM MMBench V1.1 EN - Attribute Recognition | 95.5 | Accuracy (%) | 86.2 |
| OpenVLM MMBench V1.1 EN - Image Scene | 89.6 | Accuracy (%) | 81.3 |
| OpenVLM MM-Vet - Language Generation | 57 | Score | 78.8 |
| OpenVLM MM-Vet - Knowledge | 53.9 | Score | 74.6 |
| OpenVLM MM-Vet - Recognition | 63.9 | Score | 71.6 |
| OpenVLM MMMU - Management | 60 | Accuracy (%) | 71.4 |
| OpenVLM MM-Vet - OCR | 69.7 | Score | 69.6 |
| OpenVLM OCRBench - Key Information Extraction | 84 | Correct Answers (%) | 69.5 |
| OpenVLM MMBench V1.1 EN - Action Recognition | 88 | Accuracy (%) | 69.3 |
Interactive version: theaggregate.ai/model?slug=qtunevl1-2b · How It Works · Data refreshed daily, snapshot 2026-11-08.