v4: Longer training time, same dataset, very similar settings. Produces higher detail anime style images. Use at lower strength, may be overfit. ~8.2 hours of training.





Activation: blowjob
Concept: fellatio
SDXL has been lacking some decent ???? anime LoRAs. Out of curiosity I decided to try out SDXL LoRA training with Prodigy since I got a better GPU, and I decided to give this concept a shot. I put together a dataset of training images and wrote some scripts to help sort through them, and these are some of my first successful attempts at a LoRA.
This is a WIP project and I hope to release better models soon, but give these a shot and post some of your results. Leave a heart/rating if you like this model!
Trained on: kohakuXL_nyan
v4: Longer training time, same dataset, very similar settings. Produces higher detail anime style images. Use at lower strength, may be overfit. ~8.2 hours of training.
1- النموذج المعاد النشر هو فقط لأغراض التعلم والتبادل والمشاركة، حقوق النشر والتفسير النهائي تعود للمبدع الأصلي.
2- إذا رغب المبدع الأصلي للنموذج في استلام نموذجه، يرجى التواصل مع موظفي SeaArt AI عبر القنوات الرسمية للتوثيق والاعتماد. نحن نلتزم بحماية حقوق كل مبدع. انقر للاستلام
