A simple implementation of a CLIP that splits up an image into quandrants and then gets the embeddings for each quandrant
By chatting or signing in you agree to the Terms and chat-message logging (revocable in History).