Video Question Answering on NA-VQA (Short)
54.29Comprehension ScoreClaude Sonnet 4.5
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Claude Sonnet 4.5Type=Proprietary, Use Descr.=true2026.03 | 54.29 | 59.1 | 54.01 | 51.82 | |
| GPT-120B OSSType=Proprietary, Use Descr.=true2026.03 | 45.78 | 50.14 | 43.87 | 43.71 | |
| Qwen-235BType=Proprietary, Use Descr.=true2026.03 | 43.3 | 48.31 | 44.33 | 42.9 | |
| Video-NaRAType=Fine-Tuned Model, Use Descr.=true, Frames=1282026.03 | 35.11 | 38.37 | 33.4 | 32.59 | |
| Video-NaRAType=Zero-Shot VLM, Use Descr.=true, Frames=1282026.03 | 34.26 | 36.78 | 32.38 | 31.39 | |
| Video-LLaMA3 (7B)Type=Zero-Shot VLM, Frames=1282026.03 | 33.08 | 36.38 | 29.21 | 30.1 | |
| Qwen2.5-VLType=Fine-Tuned Model, Frames=1282026.03 | 32.63 | 36.21 | 29.74 | 30.79 | |
| Qwen2.5-VL (7B)Type=Zero-Shot VLM, Frames=1282026.03 | 32.54 | 35.83 | 29 | 29.65 | |
| InternVL2.5 (7B)Type=Zero-Shot VLM, Frames=642026.03 | 32.22 | 34.66 | 32.06 | 29.65 | |
| MovieChat (7B)Type=Zero-Shot Long-Form VLM, Frames=20482026.03 | 31.52 | 35.56 | 24.84 | 29.04 | |
| VideoChat-Flash (7B)Type=Zero-Shot Long-Form VLM, Frames=30002026.03 | 30.34 | 33.27 | 24.31 | 26.35 | |
| LLaVA-OneVision (7B)Type=Zero-Shot VLM, Frames=322026.03 | 28.68 | 31.94 | 23.39 | 24.32 | |
| InternVL2 (7B)Type=Zero-Shot VLM, Frames=322026.03 | 28.39 | 29.98 | 25.83 | 25.21 | |
| VideoChatGPT (7B)Type=Zero-Shot VLM, Frames=1002026.03 | 27.82 | 31.93 | 22.93 | 21.34 | |
| Qwen2.5 Instruct (7B)Type=Zero-Shot VLM, Use Descr.=true2026.03 | 25.99 | 29.9 | 25.54 | 22 | |
| VideoLLaVA (7B)Type=Zero-Shot VLM, Frames=82026.03 | 18.62 | 21.14 | 17.23 | 14.18 |