supporing negative prompt, and using other diffusion models

ZichengDuan / TheChosenOne

Unofficial implementation of the paper "The Chosen One: Consistent Characters in Text-to-Image Diffusion Models"

237 stars 22 forks source link

supporing negative prompt, and using other diffusion models #6

Open NirDiamant opened 9 months ago

NirDiamant commented 9 months ago

Hi ZichengDuan! can you please support a negative prompt? it may help to converge to the wanted character - i.e. a real human being and a CGI.

additional question: only stable-diffusion-xl-base-1.0 (among the known models) have tokenizer_2 and text_encoder_2: how can I modify the code to work with other diffusion models?

Thanks!