Text-to-Video Retrieval on MSVD (zero-shot)
83.3Recall@1InternVideo2-1B
Evaluation Results
| Method | Links | |
|---|---|---|
| InternVideo2-1BSetting=ZS2026.05 | 83.3 | |
| Gemini Embedding 2Setting=ZS2026.05 | 77.9 | |
| Gemini Embedding 2Setting=ZS2026.05 | 77.1 | |
| PE-core familySetting=ZS2026.05 | 76.7 | |
| LanguageBind (CLIP-H/14)Setting=ZS2026.05 | 72 | |
| SigLIP-2 familySetting=ZS2026.05 | 67.2 | |
| OmniRetriever-7BSetting=ZS2026.05 | 66.9 | |
| X-CLIPSetting=FT2026.05 | 66.8 | |
| OmniRetriever-7BSetting=ZS2026.05 | 63.3 | |
| InternVideo2-1BSetting=ZS2026.05 | 58.1 | |
| CLIP4ClipSetting=FT2026.05 | 56 | |
| Omni-Embed-NemotronSetting=ZS2026.05 | 55.8 | |
| LanguageBind (CLIP-H/14)Setting=ZS2026.05 | 53.9 | |
| X-CLIPSetting=FT2026.05 | 50.4 | |
| PE-core familySetting=ZS2026.05 | 50.4 | |
| UMT-L#Pairs=17M, Zero-shot=true2023.03 | 49.9 | |
| Omni-Embed-NemotronSetting=ZS2026.05 | 49.2 | |
| UMT-L#Pairs=25M, Zero-shot=true2023.03 | 49 | |
| SigLIP-2 familySetting=ZS2026.05 | 49 | |
| CLIP4ClipSetting=FT2026.05 | 46.2 | |
| UMT-L#Pairs=5M, Zero-shot=true2023.03 | 44.4 | |
| InternVideo#Pairs=646M, Zero-shot=true2023.03 | 43.4 | |
| UMT-B#Pairs=25M, Zero-shot=true2023.03 | 42.3 | |
| UMT-B#Pairs=17M, Zero-shot=true2023.03 | 41.4 | |
| CLIP4Clip#Pairs=400M, Zero-shot=true2023.03 | 36.2 | |
| UMT-B#Pairs=5M, Zero-shot=true2023.03 | 36.2 |