First concept lora.
Please set the weight exactly at 0.5, higher will lead to bad image quality.
suugested prompts:
clothed male and nude female, bride veil, suit, white thighhighs
Description
suggested weight :0.5
FAQ
Comments (4)
Funny but pretty incontrollable, as feared.
It works until you leave the prompt generic. When you try to add something more specific, the entire stuff will mess up, and you'll get single people, or mixed gender people. Part of the problem is the notorius incabability by Stable Diffusion to handle two different subjects...
Have you tried it with regional prompter?
@firemanbrakeneck I activated the regional prompter to generate the input image for trainning this LoRA. One of purpose of this LoRA is to help user generate image without using unfamiliar extensions.
I will try other p;rompts to see what will happen. I tried centaur, some of the result are good. for mixed gender people, you can add some names of celebrities in prompts.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.