Recommend using a weight of 1.3.
v0.6 is more steps
v0.6.5 is more ranks
896x896 training
v0.5 not recommended
I trained the model incorrectly, something needs to be changed... and i don't know what.
UPD: мне пока лень, LoRA заморожена на неопределённый срок.
Description
The first attempt, too few steps or something else. Because of this, it is necessary to increase the weight of the LoRA itself, and because of this, the image deteriorates.
FAQ
Details
Downloads
269
Platform
SeaArt
Platform Status
Available
Created
7/8/2024
Updated
7/23/2024
Deleted
-
Trigger Words:
ABOBA
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.