Video Question Answering on ViTexQA
73.5Average ScoreFrameThinker
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| FrameThinkerSize=8B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 73.5 | 67.4 | 87.2 | 75.9 | 62.3 | |
| Gemini-3 ProMax_Frames=128, Resolution=512*512, Zero-shot=true2026.06 | 71.6 | 70.9 | 70.3 | 58.9 | 59.2 | |
| MiniCPM-V4.5Size=8B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 70.1 | 48.6 | 77.4 | 51.1 | 60.1 | |
| Qwen3-VLSize=8B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 66.7 | 63.7 | 61.7 | 73.6 | 53.4 | |
| InternVL3Size=8B, Max_Frames=64, Resolution=448*448, Zero-shot=true2026.06 | 64.2 | 52.1 | 85.5 | 60 | 52.3 | |
| Qwen2.5-VLSize=7B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 63.5 | 64.4 | 63.9 | 73.3 | 58.5 | |
| Qwen3-VLSize=4B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 62.3 | 62.1 | 59 | 71.7 | 57.7 | |
| Seed-1.6Max_Frames=128, Resolution=512*512, Zero-shot=true2026.06 | 61.1 | 63.4 | 60.2 | 54.9 | 56.8 | |
| Qwen3-VLSize=2B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 57.2 | 56.9 | 56.9 | 69.1 | 52.4 | |
| Qwen2.5-VLSize=3B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 55.4 | 58.1 | 53.3 | 63.2 | 49.4 | |
| VideoLLaMA3Size=7B, Max_Frames=64, Resolution=336*336, Zero-shot=true2026.06 | 52.4 | 54.5 | 62.1 | 58.5 | 49.6 | |
| GPT-4oMax_Frames=128, Resolution=512*512, Zero-shot=true2026.06 | 51.6 | 55.2 | 41.4 | 51.9 | 46.6 | |
| MiniCPM-V2.6Size=7B, Max_Frames=64, Resolution=448*448, Zero-shot=true2026.06 | 50.7 | 55.6 | 51.7 | 57 | 38.2 | |
| LongVUSize=3B, Max_Frames=768, Resolution=448*448, Zero-shot=true2026.06 | 43.1 | 45.3 | 47.5 | 46.5 | 36.4 | |
| Keye-VLSize=8B, Max_Frames=768, Resolution=336*336, Zero-shot=true2026.06 | 39.9 | 54.9 | 47.1 | 49.1 | 35.3 | |
| LLaVA-VideoSize=7B, Max_Frames=16, Resolution=336*336, Zero-shot=true2026.06 | 36 | 58.3 | 46.3 | 41 | 37.6 |