Multiple Choice Video-QA on Perception (test)
63.4AccuracyVideoChat2-HD-F16
Evaluation Results
| Method | Links | |
|---|---|---|
| VideoChat2-HD-F16ViEncoder=InternVideo2_S3-1B, LLM=Mistral-7B, Frames=162024.03 | 63.4 | |
| VideoChat2-HDViEncoder=InternVideo2_S3-1B, LLM=Mistral-7B2024.03 | 60.1 | |
| Divot-LLMLLM size=7B, Video-Gen=true2024.12 | 58.3 | |
| PLLAVALLM size=34B, Video-Gen=false2024.12 | 58.1 | |
| VideoLLaMA2LLM size=72B, Video-Gen=false2024.12 | 57.5 | |
| Gemini 1.0 Ultra2024.03 | 54.7 | |
| VideoChat2ViEncoder=InternVideo2_S3-1B, LLM=Mistral-7B2024.03 | 53 | |
| VideoLLaMA2ViEncoder=CLIP-L-336, LLM=Mistral-8*7B2024.03 | 52.2 | |
| VideoLLaMA2ViEncoder=CLIP-L-336, LLM=Mistral-7B2024.03 | 51.4 | |
| VideoLLaMA2LLM size=7B, Video-Gen=false2024.12 | 51.4 | |
| Gemini 1.0 Pro2024.03 | 51.1 | |
| Gemini 1.0 ProVideo-Gen=false2024.12 | 51.1 | |
| LLaVA-Next-VideoViEncoder=CLIP-L, LLM=Vicuna-7B2024.03 | 48.8 | |
| LLaVA-NeXT-VideoLLM size=7B, Video-Gen=false2024.12 | 48.8 | |
| Video-LaVITLLM size=7B, Video-Gen=true2024.12 | 47.9 | |
| VideoChat2LLM size=7B, Video-Gen=false2024.12 | 47.3 | |
| LLaMA-VIDLLM size=7B, Video-Gen=false2024.12 | 44.6 | |
| Video-LLaVALLM size=7B, Video-Gen=false2024.12 | 44.3 |