This model creates breasts massage scene. Works for both realistic and anime images. If faces generates poorly follow this guide.
FLUX
There are 2 ways of prompting the model. You can use tags or write a more humanlike description. From what I tested the description gives better results, however using only tags also works. Examples:
breast massage, a photo of a beautifull woman, pov, breast grab, medium breasts, nude, lying, on back, shiny skin, looking at viewer <lora:breastmassagev1_FLUX:1>or
A photo from man's pov massaging woman's breasts by grabbing them with both hands. The woman in her early 20s is lying on her back on the ground completely filled with colorful flowers. She is looking straight at the camera with her wide open eyes. Her face has extremely detailed skin texture. Her skin is oiled which gives it a shining effect. She has large breasts, ginger hair and medium length hair. <lora:breastmassagev1_FLUX:0.7>Unfortunately this model affects faces but lowering the lora's weight helps.
Recommended Lora weight: around 0.5-1.0
Prompts: breast massage, 1girl, 1boy, pov, breast grab, grabbing, large breasts, medium breasts, small breasts, nude, lying, on back, shiny skin, looking at viewer
SD1.5/SDXL
Prompts: breast massage
Other usefull prompts: 1girl, 1boy, pov, breast grab, grabbing, large breasts, medium breasts, small breasts, nude, lying, on back, shiny skin, looking at viewer
Recommended Lora/LoCon weight: around 1.0
You can vote or give your idea for next trained model here! I'm gonna try to train most popular ones.
If you liked my locon consider buying me a Coffee or giving this model 👍 and posting your creations😊
Description
redone the database, slightly diffrent training settings, V2 works more "stable" but is less flexible
FAQ
Comments (6)
I uploaded my model to Civit less then 2 hours ago and someone already reprinted it to Tensor... XD. I need to claim it, so it'll be on my profile probably in a day.
for the looooove of boobs, they should really get this guy a nobel prize 🍈🍈
Haha, thanks!
hello, I want to ask 2 questions
1- that I can't find the answer to, I'm quite new to this. I'm trying to create a model for myself, but in every image I reach a different woman. In your images all women are the same. How do you ensure this?
2-when I download this file the extension shows .safetensors How can I use this file?
1. The faces are similar because I used similar prompts in all the showcase images
2. Place it in the Lora/Lycoris folder. For A1111 Web UI (1.5 or newer) it's /webui/models/Lora
Works great at 512x512. Once you stretch past that though, you start getting totem/centipede glitches. Possible to work past with some creative negative prompting, but it takes some time.