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
4.48k stars 293 forks source link

The result using demo script cannot be aligned with the result generated using webUI #318

Open Haowenyue opened 3 months ago

Haowenyue commented 3 months ago

For the same input, the same seed, and the same model weights, the results we get with the ip-adapter-faceid inference script are quite different from the results we get with the webU.

xiaohu2015 commented 3 months ago

yes, it may be different, webui and diffusers have some difference

Haowenyue commented 3 months ago

yes, it may be different, webui and diffusers have some difference

I wonder if there is a way to align the results of the inference script with the results of the webUI, thanks!