Home Artists Posts Import Register

Downloads

Content

What have you been eating? Yeast?🍒🍈

🖼️ See some examples 🖼️

⚖️LoRA prompt weight: 0.8~1
⚙️~90% effectiveness
🏫Model used for training: personal modded model

I can't believe that the image above is a single generation. This LoRA is quite flexible considering the themes that can be used in image generation. It's not perfect yet, but you'll be able to create some very nice images.

🔍Recomendations:

  • CFG ~7
  • Steps >=35
  • DPM Family

🖼️Possible resolutions:

  • 512 x 768: 2 girls, vertical
  • 640 x 640: 2 girls, horizontal
  • 840 x 640: 3 girls, horizontal
  • 960 x 640: 3 or 4 girls, horizontal

I recommend start your prompts with:

[# of girls], side by side shot, [woman main characteristics], hyperflat small chest, busty, [clothing, viewpoint]...

LoRA details and tendencies (+resolution):

  • I achieved good results with these neg embeddings: CyberRealistic_Negative-neg and FastNegative. But you can use normal neg prompts.
  • The generated images may have a contrast problem between them, where one of them is clearer than the other. I'll fix this in the future.
  • Because of the size of the faces in the generations, they can sometimes break, use hires to solve this.

Inside the folder I left some example images, you can use this on your prompts.

Comments

No comments found for this post.