Long-context video understanding on LongVideoBench (test)
66.7AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4ocategory=Proprietary2024.10 | 66.7 | |
| GPT-4o2024.10 | 66.7 | |
| GPT-4o2024.12 | 66.7 | |
| ARIAactivated_params=3.9B (25.3B), category=Open-source2024.10 | 65.3 | |
| ARIA2024.10 | 65.3 | |
| Gemini-1.5-Procategory=Proprietary2024.10 | 64.4 | |
| Gemini-1.5 Pro2024.10 | 64.4 | |
| LLaVA-OneVision-72Bactivated_params=72B, category=Open-source2024.10 | 63.2 | |
| Gemini-1.5-Flashcategory=Proprietary2024.10 | 62.6 | |
| Gemini-1.5 Flash2024.10 | 62.4 | |
| Qwen2-VL-72Bactivated_params=72B, category=Open-source2024.10 | 61.7 | |
| GPT-4V2024.10 | 60.7 | |
| InternVL2-40Bactivated_params=40B, category=Open-source2024.10 | 60.6 | |
| GPT-4o minicategory=Proprietary2024.10 | 58.8 | |
| GPT-4o mini2024.10 | 58.8 | |
| GPT-4o mini2024.12 | 58.8 | |
| NVILANumber of Parameters=8B, #Frames=2562024.12 | 58.7 | |
| Qwen2-VL-7Bactivated_params=7B, category=Open-source2024.10 | 56.8 | |
| Qwen2-VLNumber of Parameters=8B, #Frames=2fps2024.12 | 56.8 | |
| MiniCPM-V-2.6activated_params=8B, category=Open-source2024.10 | 55.7 | |
| InternVL-Chat-V1.5activated_params=26B, category=Open-source2024.10 | 51.7 | |
| Video-XLNumber of Parameters=7B, #Frames=20482024.12 | 51.3 | |
| Idefics2activated_params=8B, category=Open-source2024.10 | 49.4 | |
| Pixtral-12Bactivated_params=12B, category=Open-source2024.10 | 47.4 | |
| Pixtral-12B2024.10 | 47.4 | |
| Llama3.2-11Bactivated_params=11B, category=Open-source2024.10 | 45.7 | |
| Llama3.2-11B2024.10 | 45.7 | |
| LLaVA-NeXT-VideoNumber of Parameters=7B, #Frames=322024.12 | 43.5 |