You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'd like to suggest to implement the same approach on the ControlNet. So, let's say I want the first CN to start at step 0 with weight 0.5 , and then gradually reduce until in vanishes at last step. I could write in the prompt something like this.
< CN1 : 0.5 @ 0 , 0 @ 1 >
With this, I could do something like letting one controlNet start strong, then let another one gradually taking over instead of abruptly cut and sampler starts doing random things.
I've requested for this before at LoraCTL long ago. But I guess it is challenging to be implemented over there. cheald/sd-webui-loractl#21
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
I've been using LoraCTL plug in to control the lora from prompt. See here: https://github.com/cheald/sd-webui-loractl
I'd like to suggest to implement the same approach on the ControlNet. So, let's say I want the first CN to start at step 0 with weight 0.5 , and then gradually reduce until in vanishes at last step. I could write in the prompt something like this.
With this, I could do something like letting one controlNet start strong, then let another one gradually taking over instead of abruptly cut and sampler starts doing random things.
I've requested for this before at LoraCTL long ago. But I guess it is challenging to be implemented over there.
cheald/sd-webui-loractl#21
Beta Was this translation helpful? Give feedback.
All reactions