Tag
A model on Replicate that outputs CLIP ViT-L/14 features for text and images, allowing similarity computation between inputs.