Video Captioning on Dream-1K
48.21PrecisionCuRe
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CuReCategory=Open-source2026.07 | 48.21 | 36.26 | 41.39 | |
| Gemini 3.0 ProCategory=Proprietary2026.07 | 42.67 | 39.26 | 40.89 | |
| Tarsier2-7BCategory=Open-source2026.07 | 41.5 | 38.8 | 40.1 | |
| Tarsier-34BCategory=Open-source2026.07 | 41.4 | 32.4 | 36.3 | |
| InternVL3.5-30B-A3BCategory=Open-source2026.07 | 39.07 | 28.17 | 32.74 | |
| LLaVA-Video-7BCategory=Open-source2026.07 | 37.9 | 28.4 | 32.5 | |
| GPT-4o2026.01 | 36 | 43 | 39 | |
| Gemini1.5 Pro2026.01 | 35 | 38 | 36 | |
| Tarsier2-7B*2026.01 | 35 | 47 | 40 | |
| Qwen2.5-VL-72B-InstructCategory=Open-source2026.07 | 34.72 | 28.9 | 31.54 | |
| OwlCap-7BCategory=Open-source2026.07 | 34.1 | 35.3 | 34.7 | |
| LFSBackbone=Tarsier2-7B*2026.01 | 34 | 48 | 40 | |
| Qwen3-VL-30B-A3B-InstructCategory=Open-source2026.07 | 33.35 | 38.32 | 35.66 | |
| Seed 2.0 ProCategory=Proprietary2026.07 | 32.97 | 50.8 | 39.99 | |
| GPT-4V2026.01 | 30 | 41 | 34 | |
| Qwen3-VL-235B-A22B-InstructCategory=Open-source2026.07 | 29.48 | 40.77 | 34.22 | |
| VideoChat22026.01 | 23 | 31 | 27 | |
| MiniGPT-4V2026.01 | 22 | 26 | 24 | |
| PLLaVA-34B2026.01 | 22 | 38 | 28 | |
| LLaVA-NeXT-Video2026.01 | 21 | 36 | 26 | |
| Video-LLaVA2026.01 | 16 | 28 | 20 |