Can you provide a code for Text - Image similarity?
We currently provide text/image similarity for the LiT-tuned checkpoint here.
The two models share the same text encoder, right?
· Sign up or log in to comment
