Trained on 100 images.
4 of them are of Android 17 and 18 being absorbed from the source.
The other 96 are various AI generated, Cellvore pics on this site lol.
1.0 to 1.2 weight should be good enough.
I have no idea how people on Pixai and Civitai trained their cellvore LoRAs. So this is a first time for me. Well not really, my internal vore view LoRA was made with just internal shots in which the victims can be interchanged with whatever. So perhaps the same can apply to this LoRA.
Most of the images had Semi-perfect Cell's tail, so don't be surprised if you end up getting Imperfect Cell's tail.
The XXX can be whatever absorption victim you want.
I am still figuring this out. If you've managed to come up with better trigger words than me, with prompt helper on or off, and it resulted in better images, please feel free to share them in the comments below.
And you may want to put prompts like "head" and "hair" in the negatives just in case.
These prompts can help:
"the girl was caught by the sucker,while the absorption started"
"hair spitts out of the sucker"
