NOW ON FANVUE, CREATING FULL LENGTH ADULT CONTENT WITH SOME OF YOU'RE FAVORITES: https://www.fanvue.com/zikickix
UPDATE: Now trained with the Base model including merging with the base model. Replaced some images from 1.5 with Anima images, Z-Turbo images for higher quality images.
This uses my trained Zi lora/checkpoints from Illustrious mixed in with Anima Base!
Works very well with natural language.
YOU NEED:
qwen base encoder & qwen image vae
All gallery images should have metadata so you can drag and drop to see the exact things you need.
SETUP:
Steps: 30 - 50
CGF: 4 - 5
Sampler: Eular/Eular A/res 2s (haven't tested others)
Description
Uses the dataset from Illustrious with modified tagging.
FAQ
Comments (1)
Wacky text compositions probably worsened from the base preview3, but it's better at some other sillies, and if you want a very nice shiny refiner then this does a good job.
Posted a couple of images with deliberately distorted proportions through Unholy Proompting to stress it some, did a good job.