Stable diffusion thread

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,732
Reaction score
3,946
Ok so TensorRT is no longer working now because it requires too much vram.
Cannot export dynamic engine due to failed onnx conversion.
 

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,732
Reaction score
3,946
Ok so I finally upgraded to the Forge version of SD WebUI.
It does seem that the upscaling process is much faster now.
 

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,732
Reaction score
3,946
Tested the new AnimateLCM-SVD-xt-1.1 model.
Can generate 25 frames video in less than 30 seconds on my local machine (excluding model loading time).
Use cfg 1.0 & 4 steps.
But the success rate is very low for 768x768 and 576x1024 sizes.
 
Last edited:

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,732
Reaction score
3,946
I managed to get DynamiCrafter working nicely after using Kijai's ComfyUI wrapper.
Success rate is much higher than SVD, so no more generating a huge number of unwanted garbled animations.
However, prompt adherence is not exactly 100%.
Generation performance is fast for 16 frames on my local machine, takes about 4 minutes.
Using 30 steps and cfg 7.5.
 
Last edited:
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ Forums. Forum members and moderators are responsible for their own posts. Please refer to our Community Guidelines and Standards and Terms and Conditions for more information.
Top