Replies: 1 comment
-
|
I tried several fresh builds starting from 0.3.45. Installed minimum custom nodes packages, replaced torch 2.7 cu128 with torch2.8 cu128, replaced packages like opencv-python-headless and opencv-contrib-python, numpy. Nothing helps. :( |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
I open workflow from image generated in late august this year, run the queue and get different results. Why?
Sampler is DPMPP_2M_SDE_GPU, scheduler KARRAS. I thought developers have some refs to compare generation results, no?
Some workflows from previously generated images give more close results, some workflows generate exactly the same results, but some generate very much different results.
Thanks for your suggestions.
Beta Was this translation helpful? Give feedback.
All reactions