tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: '-'
output:
url: >-
https://wimg.rule34.xxx//images/881/e529479ad5425100f9e225c4794b3483.jpeg
- text: '-'
output:
url: >-
https://wimg.rule34.xxx//images/2271/5fda792f4c4ddbef663e00dec5510b44.jpeg
- text: '-'
output:
url: >-
https://wimg.rule34.xxx//images/2271/5fda792f4c4ddbef663e00dec5510b44.jpeg
base_model: kitty7779/ponyDiffusionV6XL
instance_prompt: soiling
license: openrail
Soiling(Pony)
- Prompt
- -
Model description
CONTENT WARNING: SCAT
CivitAI account in case you didn't come from there.(https://civitai.com/user/DarkModeOP)
Ko-Fi if you are really generous.(I don't need the money, I have a comfy life.) https://ko-fi.com/darkmodeop
Here is a LoRA model that is too spicy for CivitAI
Example images here because I have no idea what Hugging Face's rules for this kind of NSFW content is. https://lensdump.com/a/Sv1VC
Hello again, I plan to do some refinements but this is a good start. This was trained with "overflow" for the look of leaking but for my model I think overflow is associated with cum_overflow so it produced strange results.
Example Prompt: <lora:SoilXL150-Pony:0.99> 1girl, dark_skin, panties, soiling, blush, from_behind
Helpful words: overflow(Maybe)
Update(08/23/2024):
Uploaded a newer version that supports soiled diapers but due to a training issue doesn't do heavily soiling. On one hand that means smaller messes overall so I thought it might be worth uploading.
Prompting is pretty much the same.
New Gens (https://lensdump.com/a/nwCro)
Example Prompt: <lora:SoilXLV2:0.99> 1girl, dark_skin, panties, soiling, blush, from_behind
Helpful words: diaper
Trigger words
You should use soiling to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.