Update `image_encoder_path` to a public CLIP one
#15
by multimodalart HF Staff - opened
Hi there, I'm proposing this PR to correct the IP Adapter Plus demo.
However, I'm not sure that is the correct CLIP model, as the correct CLIP model is not referred to anywhere
Happy to update the PR if openai/clip-vit-large-patch14 should be the correct model (to match the model used to train SD1.5)
Looks correct to me, it's pointing to the CLIP Vision model on the main IPAdapter Repo here - https://huggingface.co/h94/IP-Adapter/tree/main/models/image_encoder
yes, thanks a lot
h94 changed pull request status to merged