If you like the model, please add image to the model page + Add Post and leave a review, it really helps. The buzz makes it easier for me to make improvements/train new models.
ZIT:
Prompting: gagged with a ball gag
PONY V3:
Prompting: large ball gag, (color) ball gag, gagged
Much better version.
I noticed most of my generation buzz comes from the pony version of this model, which has allowed me to train more models. This is a better version as thanks for using my models. should be able to control the color more consistently, plays better with other loras and gives cleaner outputs.
PONY V2:
Prompting: b4llg4g, large ball gag, gagged,
This is a much cleaner version of the pony model that won't introduce blurriness like V1
FLUX:
Prompting: (Subject) gagged with large (color) b4llg4g in mouth with rivets and leather straps
Capable of decent nudity without support from other LoRas
Can do medium/normal size by not mentioning "large" and saliva with "saliva. (Better at full strength)
If trying to do non-realistic style and having issues, either increase strength of other lora or lower this lora's strength to 0.85
Description
Pony version, I don't have much experience training pony models, sometimes the results come out blurry, especially if it's not a close-up. If anyone is experienced with pony and doesn't mind sharing good settings for the on-site trainer, I would really appreciate it and will retrain the model. I don't feel like the dataset is the issue, I have >50 high quality images in different contexts and angles. With this model, I can take the best images between both versions and increase the data set to 100 images.