HawkVL-2B: benchmark results
Provider: Various. Released 2025-03-03.
Aggregate Rating 1366 ± 30, rank #7283 of 10916 rated models, from 148 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM OCRBench - Text Recognition | 89.67 | Correct Answers (%) | 84.2 |
| OpenVLM MMMU - Diagnostics and Laboratory Medicine | 46.7 | Accuracy (%) | 82.3 |
| OpenVLM OCRBench - Handwritten Mathematical Expression Recognition | 62 | Correct Answers (%) | 69.2 |
| OpenVLM AI2D - Eclipses | 90.3 | Accuracy (%) | 64 |
| OpenVLM MMMU - Computer Science | 56.7 | Accuracy (%) | 62.7 |
| OpenVLM MMMU - Energy and Power | 43.3 | Accuracy (%) | 62 |
| OpenVLM AI2D - Volcano | 87.5 | Accuracy (%) | 56.9 |
| OpenVLM MMMU - Marketing | 50 | Accuracy (%) | 56.2 |
| OpenVLM MMMU - Mechanical Engineering | 36.7 | Accuracy (%) | 54.8 |
| OpenVLM MMMU - Art Theory | 70 | Accuracy (%) | 52.7 |
| OpenVLM OCRBench | 76.3 | Score (normalized) | 51.5 |
| OpenVLM MMMU - Materials | 33.3 | Accuracy (%) | 49.5 |
Interactive version: theaggregate.ai/model?slug=hawkvl-2b · How It Works · Data refreshed daily, snapshot 2026-11-08.