Multi-discipline Video Understanding on VideoMME
71.9AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o2024.08 | 71.9 | |
| LLaVA-OneVision-72BModel Scale=72B2024.08 | 66.2 | |
| GPT-4VVersion=V-Preview2024.08 | 59.9 | |
| LLaVA-OneVision-7BModel Scale=7B2024.08 | 58.2 | |
| LLaVA-OneVision-0.5BModel Scale=0.5B2024.08 | 44 |