Text-to-Image Retrieval on Flickr30k (Recall@1)
89.9Recall@1Voyage-3.5-multimodal
Evaluation results
| Method | Links | |
|---|---|---|
| Voyage-3.5-multimodalModality=V/I/T2026.05 | 89.9 | |
| Gemini Embedding 2Modality=V/A/I/T2026.05 | 89.1 | |
| Amazon Nova MMEModality=V/A/I/T2026.05 | 81.6 | |
| multimodalembedding@001Modality=I/T2026.05 | 81.4 | |
| TULIPBackbone=ViT-B/16, Evaluation Protocol=zero-shot2026.07 | 35.2 | |
| HyFL-CLIPBackbone=ViT-B/16, Evaluation Protocol=zero-shot2026.07 | 35.2 | |
| HiMo-CLIPBackbone=ViT-B/16, Evaluation Protocol=zero-shot, Implementation Source=our implementation2026.07 | 34.02 | |
| FineLIPBackbone=ViT-B/16, Evaluation Protocol=zero-shot, Implementation Source=our implementation2026.07 | 33.85 | |
| Long-CLIPBackbone=ViT-B/16, Evaluation Protocol=zero-shot, Implementation Source=author-provided checkpoints2026.07 | 33.16 |