search

tencent-ailab / IP-Adapter

The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.
Apache License 2.0
5.02k stars 325 forks source link

Which CLIP model do you use? #25

Closed haofanwang closed 1 year ago

haofanwang commented 1 year ago

I notice that you provide image encoder on your own space, is it different from the models released by openai?

xiaohu2015 commented 1 year ago

@haofanwang hi, for IP-Adapter of SD 1.5, we use OpenCLIP-ViT-H-14. But for IP-Adapter of SD xl, we use OpenCLIP-ViT-bigG-14. For your convenience, we have also uploaded a copy in our model space.