Sign up for free and get 10,000 credits instantly

Starter credits for new creators, added to your account on the day you sign up.

Notice: PixAI iOS App Temporarily Unavailable on the App Store — Dear PixAI users, The PixAI iOS app is currently unavailable for search and download on the App Store. We are actively w…

View details

LoRA Pengguna

SD

Doggystyle from side view

#Wanita matang#1 perempuan
Doggystyle from side view - AI Model cover image
Dimuat naik pada
18 Mac 2024, 4:41 PG
Kegunaan
2.4k
Ulasan
Tiada ulasan
Kata Pencetus

doggy style

Kebenaran
  • Allow image generation & sharing
  • Benarkan pengguna memuat turun model anda
  • Kegunaan komersial

Recommended LoRAs

Keterangan

Versionv0.2:Greatly improve the quality of imagesGreatly ruduce the posibility of no boyGreatly reduce the posibility of generiting fewer leg and fewer arm girl and other anatomy bodyGreatly improve of this Lora that can't change clothes or hair color.Problems:Fewer arm problem is still existingUsing too many prompt to decorate girl's drees may generating bad anatomy bad or no man(only penis)The color of socks is hard to change or mixed with hair colorThis is my first ever loRA,trained using Anythingv4.5,because lack of images ol this sex position,I only trained it with 2 images that accidently generated by Threesome Lora.This sex position is hard to find in internet or find this sex position but have bad quality,However,this loRA based on 2 images will generated this position often,so,version will update very soon.Weight suggest at 0.6-0.8(mixed with other Lora should test by yourself)By the way.l am a fresh,this mod have many problems that I don't know why it happened .if you wanna improve this loRA ,just do it.Known problems:no boy or no sexbad anatormyconfused image that is lllagible(forgive my poor english)Solution(cum on body,cum in pussy,(1boy,1girl),sex from behind,implied sex,doggystyle),put it at beginning willapparently reduce the probability of thisgenerate another image or using EasyNegative and ng_deepnegative isa efficient way to improve it.change sampler to DPM SDE karrasHow to useput it at models\Lora,and add this lora to your prompt

Komen

Karya oleh Doggystyle from side view