VideoLLaMA2.1-7B: benchmark results
Provider: Other.
Aggregate Rating 1478 ± 50, rank #2689 of 10916 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MA-Bench - Coarse-Grained Recognition | 34.3 | Accuracy (%) on the coarse-grained (body-level) micro-action | 90.5 |
| BBC News Video Captioning - Text Similarity | 29.01 | Text similarity (cosine, -1 to 1, times 100) between all-Min | 57.1 |
| MA-Bench (Closed-Ended) | 42.57 | Accuracy (%) averaged over the six closed-ended tasks (coars | 47.6 |
| MA-Bench - Fine-Grained Recognition | 26.8 | Accuracy (%) on the fine-grained (action-level) micro-action | 42.9 |
| MMVU | 39.8 | Hidden-test overall accuracy, chain-of-thought (%) | 40 |
| MA-Bench (Open-Ended) | 0.17 | GPT-4o judge score (0 to 5) averaged over the six open-ended | 38.1 |
| BBC News Video Captioning - Entity Fidelity | 5.3 | Entity Fidelity Score (0-1, times 100): per-clip harmonic me | 28.6 |
Interactive version: theaggregate.ai/model?slug=videollama2-1-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.