Short Video Question Answering on Perception Test (val)
74.3AccuracyLLaVA-Video-72B
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaVA-Video-72BModel Group=Open-source (>10B), Parameters=72B, Training set observed=true2025.01 | 74.3 | |
| Tarsier2-7BParameters=7B, Training set observed=true2025.01 | 71.6 | |
| Previous SOTACitation=[72], Training set observed=true2025.01 | 70 | |
| LLaVA-Video-7BModel Group=Open-source (<=10B), Parameters=7B, Training set observed=true2025.01 | 67.9 | |
| LLaVA-OV-72BModel Group=Open-source (>10B), Parameters=72B2025.01 | 66.9 | |
| Qwen2-VL-72BModel Group=Open-source (>10B), Parameters=72B2025.01 | 66.5 | |
| Tarsier-34BModel Group=Open-source (>10B), Parameters=34B2025.01 | 60.4 | |
| LLaVA-OV-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 57.1 | |
| Tarsier-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 53.9 |