Manuel Schmid

@alexamore the issue most certainly is that your GPU is throttled by constantly having to outsource to RAM/swap instead of keeping the SDXL model in VRAM, which strangles performace. You...

Appendix: Above comment was writen under the assumption that you've already exhausted the options to update your driver to the latest version as well as set up swap correctly. Thank...

Maybe one last thing: can you please elaborate what you mean by "crashes everything"? Is your PC crashing or Fooocus, are there any logs in Fooocus console logs or the...

Crazy, never thought about the NVMe being a problem. Thank you for sharing!

@ItsNoted thank you for the post, i already read it a few days ago and with Dreamshaper XL also XL resolutions work really well. Making it easier to use by...

@IamTirion oh, i didn't know about that, thank you for the information! This might lead to us having to mention/reference this information somewhere when we add support for one of...

Models for the chosen preset (if given, also default) are already downloaded on application startup in https://github.com/lllyasviel/Fooocus/blob/main/launch.py#L73-L89. Please make sure to add all necessary model links to your config or...

@vanetreg @eddyizm this is not a bug but a limitation of RAM, mentioned in this discussion: https://github.com/lllyasviel/Fooocus/discussions/74#discussioncomment-7966878 Dragging images down from generated results to input image doesn't work on Colab...

@vanetreg yeah, sorry for getting off-topic. I couldn't find anything for Replit, but for replicate.com though ^^ @eddyizm last off-topic comment: no, input image is not working on Colab (Free),...

@vanetreg did you find a solution? I mean, basically it's a docker setup with a container, right? You might be able to build the container with the help of changes...