Metric: PC-VID (higher is better)
| # | Model↕ | PC-VID▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | GPT-4o | 66.75 | No | GPT-4o System Card | 2024-10-25 | - |
| 2 | LLaVA-Video-7B | 59 | No | Video Instruction Tuning With Synthetic Data | 2024-10-03 | - |
| 3 | LLaVA-OneVision-7B | 47.25 | No | LLaVA-OneVision: Easy Visual Task Transfer | 2024-08-06 | Code |
| 4 | Qwen2.5-VL-7B | 46.5 | No | Qwen2.5-VL Technical Report | 2025-02-19 | Code |
| 5 | InternVL2.5-26B | 21.75 | No | Expanding Performance Boundaries of Open-Source ... | 2024-12-06 | Code |
| 6 | Qwen2-VL-7B | 16.25 | No | Qwen2-VL: Enhancing Vision-Language Model's Perc... | 2024-09-18 | Code |
| 7 | mPLUG-Owl3-7B | 13.5 | No | mPLUG-Owl3: Towards Long Image-Sequence Understa... | 2024-08-09 | Code |
| 8 | InternVL2.5-8B | 5.25 | No | Expanding Performance Boundaries of Open-Source ... | 2024-12-06 | Code |
| 9 | LongVA-7B | 3.75 | No | Long Context Transfer from Language to Vision | 2024-06-24 | Code |