This model is trained from japan av covers - Harem Concept AV.
Let me tell you about trigger words. In here, many trigger words exist.
AV Concept : bright, cute, sexy
Place : outdoors, house, classroom, office, hospital, soapland, pool, spa
descript girl : student, massage parlor girl, nurse, teacher, race queen, gal, office lady, maid
cloth : school uniform, nurse uniform, nurse cap,sportswear, onepiece swimsuit, umbrella, bikini, lingersuit, bunny girl, bunny girl hairband, kimono, rope bondage, black latex
nudity : nude, semi nude, pubic hair, big breasts, gigantic breasts
behavior : exhibit panties,sit on chair, sit on desk
camera position : ass( => camera watching behind girls)
number of girls : 1girl, 2girls, 3girls, 4girls(4, 5, 6,...)
Description
FAQ
Comments (8)
Memo.
The animation model does not seem to be supported.
(Weight: The picture does not collapse after about 0.6, but the package no longer looks like a package.)
It is very good that you can create a nice package by inputting elements according to 1~8 in the summary field.
How do you compensate for facial detail?
I used detection detailor. This is a plugin of Stable Diffusion.
@boxak30134955 Thank you for your response.
You are truly a genius
I'm glad to hear that. Thank you!
This is a very good LoRA with excellent training! The poses and placement of hands are almost spot on.
Possible to do one with just AV cover type of poses without the text?
would love to see this for newer base models.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.








