search

CompVis / stable-diffusion

A latent text-to-image diffusion model
https://ommer-lab.com/research/latent-diffusion-models/
Other
66.51k stars 9.97k forks source link

Where can I get the one good with 256x256 #847

Open Pang-0093 opened 2 months ago

Pang-0093 commented 2 months ago

Hi there! I want to operate on 256x256 size images. I try to set the --H and --W to 256 and prompt the SDv1.4, but I usually obtain some corrupted images (and when H and W are 512, all is well then). The model card shows that the SD1.4 is pretrained on 256 and finetuned on 512. So I assume that the diffusion module is finetuned to fit 512 size more. I want to know is the SD1.4 pretrained on 256 size published? or where can I get the good one with size 256x256? Can anyone help me out? Thanks a lot!

CreamyLong commented 1 month ago

you could use this model https://ommer-lab.com/files/latent-diffusion/text2img.zip