Video-to-Text Retrieval on MSR-VTT 1K (test)
53.7R@1InternVideo2 6B
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| InternVideo2 6BRetrieval Stage=Dual, Parameters=6B, Zero-shot=true2026.01 | 53.7 | 77.5 | — | — | — | — | |
| PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 51.2 | — | — | — | — | — | |
| InternVideo2V-Enc Params=1.0B, Video Data=102M, Resolution=224, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 50.9 | — | — | — | — | — | |
| PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 50.5 | — | — | — | — | — | |
| Side4Video E/14Pretrain=LAION-2B, Fine-tuning Protocol=Frozen backbone, Backbone=ViT-E/142023.11 | 50.4 | 77.4 | 86 | 1.5 | 7.1 | — | |
| PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 50.3 | — | — | — | — | — | |
| PE-LV-Enc Params=0.3B, Video Data=22M, Resolution=336, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 50.1 | — | — | — | — | — | |
| Side4Video L/14Pretrain=CLIP-400M, Fine-tuning Protocol=Frozen backbone, Backbone=ViT-L/142023.11 | 50 | 77.1 | 85.9 | 1.5 | 7 | — | |
| PE-GV-Enc Params=1.9B, Video Data=22M, Resolution=448, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 49.9 | — | — | — | — | — | |
| PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 49.6 | — | — | — | — | — | |
| PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 49.4 | — | — | — | — | — | |
| Cap4VideoPretrain=CLIP-400M, Fine-tuning Protocol=Full Fine-tuning2023.11 | 49 | 75.2 | 85 | 2 | 8 | — | |
| LanguageBindV-Enc Params=0.3B, Video Data=10M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 48.7 | — | — | — | — | — | |
| PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 48.4 | — | — | — | — | — | |
| VIRTUE v1 7BRetrieval Stage=Dual, Parameters=7B, Zero-shot=true2026.01 | 47 | 69.9 | — | — | — | — | |
| Side4Video B/16Pretrain=CLIP-400M, Fine-tuning Protocol=Frozen backbone, Backbone=ViT-B/162023.11 | 46.6 | 75.8 | 84.3 | 2 | 7.9 | — | |
| VIRTUE v2 7BRetrieval Stage=Dual, Parameters=7B, Zero-shot=true2026.01 | 46.3 | 70.7 | — | — | — | — | |
| CM AdapterPretrain=CLIP-400M, Fine-tuning Protocol=Frozen backbone2023.11 | 46.2 | 73.6 | 83.8 | — | 8.6 | — | |
| Side4Video B/32Pretrain=CLIP-400M, Fine-tuning Protocol=Frozen backbone, Backbone=ViT-B/322023.11 | 44.6 | 72.3 | 82.3 | 2 | 9.4 | — | |
| CLIP2VideoPretrain=CLIP-400M, Fine-tuning Protocol=Full Fine-tuning2023.11 | 43.5 | 72.3 | 82.1 | 2 | 10.2 | — | |
| CLIP4ClipPretrain=CLIP-400M, Fine-tuning Protocol=Full Fine-tuning2023.11 | 43.1 | 70.5 | 81.2 | 2 | 12.4 | — | |
| ImageBindV-Enc Params=0.6B, Video Data=3M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 42.9 | — | — | — | — | — | |
| VIRTUE-Embed 7BRetrieval Stage=Single, Architecture Type=MLLM-based, Parameters=7B, Zero-shot=true2026.01 | 42.6 | 65.8 | — | — | — | — | |
| InternVL-GFrames=8, Zero-shot=true2023.12 | 42.4 | 65.9 | 75.4 | — | — | 63.4 | |
| CaRe 7BRetrieval Stage=Single, Architecture Type=MLLM-based, Parameters=7B, Zero-shot=true2026.01 | 41.7 | 68.1 | — | — | — | — | |
| CLIP4Clip (ViT)Pretrain=CLIP-400M, Fine-tuning Protocol=Frozen backbone2023.11 | 41.1 | 68.9 | 78.7 | 2 | 12.4 | — | |
| LanguageBindFrames=8, Zero-shot=true, Temporal attention layers=true2023.12 | 40.9 | 66.4 | 75.7 | — | — | 62.8 | |
| InternVL-CFrames=8, Zero-shot=true2023.12 | 40.2 | 63.1 | 74.1 | — | — | 61.5 | |
| InternVLV-Enc Params=5.5B, Resolution=224, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 40.2 | — | — | — | — | — | |
| LanguageBindRetrieval Stage=Single, Architecture Type=CLIP-based, Zero-shot=true2026.01 | 40.1 | 65.4 | — | — | — | — | |
| UMT-LFrames=8, Zero-shot=true, Temporal attention layers=true2023.12 | 38.6 | 59.8 | 69.6 | — | — | 58 | |
| UMT-LV-Enc Params=0.3B, Video Data=25M, Resolution=224, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 37.1 | — | — | — | — | — | |
| InternVL-GFrames=1, Zero-shot=true2023.12 | 36.6 | 58.3 | 67.7 | — | — | 55.7 | |
| Long-CLIP L/14Retrieval Stage=Single, Architecture Type=CLIP-based, Backbone=L/14, Zero-shot=true2026.01 | 36.2 | 62.2 | — | — | — | — | |
| InternVL-CFrames=1, Zero-shot=true2023.12 | 35.3 | 56.6 | 66.6 | — | — | 54.6 | |
| SigLIP2-g-optV-Enc Params=1.1B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 34.2 | — | — | — | — | — | |
| SigLIP2-L/16V-Enc Params=0.3B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 31.4 | — | — | — | — | — | |
| OpenAI CLIP-LFrames=1, Zero-shot=true2023.12 | 27.8 | 49.4 | 58 | — | — | 45.7 | |
| OpenAI CLIP-LFrames=8, Zero-shot=true2023.12 | 26.6 | 50.8 | 61.8 | — | — | 48.1 | |
| InternVideoFrames=8, Zero-shot=true2023.12 | — | — | 39.6 | — | — | 40.7 |