Vintern-3B-beta: benchmark results
Provider: VinAI. Released 2024-08-15. Access: Open.
Aggregate Rating 1389 ± 34, rank #6550 of 10916 rated models, from 439 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Crowd Counting | 85 | Score (%) | 99.8 |
| OpenVLM MMT-Bench - Pixel Recognition | 80 | Score (%) | 99.3 |
| OpenVLM MMT-Bench - PLP | 54.5 | Score (%) | 98.1 |
| OpenVLM MMT-Bench - Rotated Object Detection | 77.8 | Score (%) | 96.4 |
| OpenVLM MTVQA - Vietnamese | 41 | Accuracy (%) | 96.2 |
| OpenVLM MMT-Bench - MeViS | 50 | Score (%) | 96.1 |
| OpenVLM MMT-Bench - Humanities and Social Science | 72.7 | Score (%) | 93.9 |
| OpenVLM MMT-Bench - Temporal Ordering | 40 | Score (%) | 92 |
| OpenVLM A-Bench - Major Object Recognition | 95.1 | Accuracy (%) | 91.2 |
| OpenVLM MMT-Bench - Depth Estimation | 75 | Score (%) | 89.1 |
| OpenVLM MMMU - Geography | 63.3 | Accuracy (%) | 88.9 |
| OpenVLM MMT-Bench - Temporal Sequence Understanding | 50 | Score (%) | 88.1 |
Interactive version: theaggregate.ai/model?slug=vintern-3b-beta · How It Works · Data refreshed daily, snapshot 2026-11-08.