Version Detail
Trained by Tensor
SDXL 1.0
8
UPDATED README: v4 is the best model, v2 is almost as good with a slightly different artstyle
AFTER LOTS OF TESTING: it's possible to get great gens without inpainting, but you need to do a lot of work with prompts and weights
If you have difficulty prompting, just copy prompt that you think looks closest to what you want.
README: v1 has better cv than v3 at the cost of cv artifacts, v3 has less cv artifacts but more stubburn with still ov artifacts, hopefully v4 improves on less artifacts
experimental, labeled sdxl but pony compatible,
STILL A WORK IN PROGRESS, WILL BE REFINED AS TESTED MORE
(please use it so i can get more credits to add more images to database c: ty)
all example images were generated on this lora with 1 additional artist style lora
v3 is capable of gut bulge and oral vore with additional tags belly and gut bulge, but still struggles with mouth artifacts.
trigger tags self explanatory, whether all of them work is still in question. ov head/feet/body don't seem to have trained correctly.
if you are getting heads of hair on your balls, turn off adetailer. v3 has acceptable faces with adetailer off, recommend generating with adetailer off and then finishing a good seed with inpainting mouth/cum areas and adetailer
negative prompts are for things lora tends to put in on its own if you dont want them
Project Permissions
Use in TENSOR Online
As a online training base model on TENSOR
Use without crediting me
Share merges of this model
Use different permissions on merges
Use Permissions
Sell generated contents
Use on generation services
Sell this model or merges