Spaces:

AP123

/

IllusionDiffusion

like 4.67k

Running on Zero

Hi, could you please describe the issue in a bit more detail with some images as well? Thanks!

Hello,

Appreciate the prompt response. I'm honestly not exactly sure what details I need to provide you with. If you have any particular questions, I can try to answer them. I notice the Upload UI has slightly changed and ever since this update, my image does NOT upload, and I cannot run anything as the "Run" button is greyed out.

To be more precise, my image "uploads" but it's just a white blank, and the "run" button still greyed out

same for me, even with examples when i press the "run" button, nothing happen

Same problem with me

Thanks everyone for reporting the issue and sorry for the inconvenience. Thanks to @cbensimon the issue should now be fixed. If the issue happens again, please let us know. In the meantime, enjoy the illusions! 🤗

Hey @Wauplin @cbensimon , I'm actually still having trouble running the tool... So the tool now runs with my prompt but it's not taking account of the uploaded image. Also, is the capability for me to reposition the image no longer available? I was able to resize/reposition as I desired in the previous UI, but don't seem to have the option anymore

Any help here...? ;(

Also, is the capability for me to reposition the image no longer available? I was able to resize/reposition as I desired in the previous UI, but don't seem to have the option anymore

Can it come from a Gradio version update? (cc @abidlabs maybe?)

I'm honestly not sure. Is it possible for users like myself to revert back to the previous version from the latest update? This is seriously one of my most enjoyable AI art tools I play around with and I am quite gutted I haven't been able to use it since the update. I'm also on discord if I can help out in any way on my end for trial-n-errors.

Thank you for the prompt response.

Hi @alsckd333 have you tried duplicating the Space and setting the sdk_version in the readme.md to 3.50.2 -- that should restore the previous functionality (there might be a few other changes to revert but that should generally work)

I cannot duplicate it for some reason without having to pay for the NVDIA GPU... Will the duplicating resolve all these issues 100%? If so, i don't mind trying but i was hoping to keep using the original space at no charge. @abidlabs . I love this tool so much, really really would like to work it out somehow without having to pay.

Hi @alsckd333 I just tested it and in my case, it does take into account the uploaded image as you can see here:

I think your input image has to black and white

I have made over 100 pepe AI art using this tool though... so definitely takes in other colours. That said, I did notice I can also use those default patterns too but I can't seem to generate images with my uploaded images anymore... Again appreciate your prompt response. Also howcome the ability to resize/relocate your image disappeared? That was a really cool/powerful tool that I used a lot