Sign In

DCAU Girls Style - DC Animated Universe Girls

162
960
6.6k
48
Updated: Jan 12, 2024
styledcausexyfemaledc2d
Verified:
SafeTensor
Type
LoRA
Stats
960
6,567
24
Reviews
Published
Jan 12, 2024
Base Model
SD 1.5
Training
Epochs: 15
Usage Tips
Clip Skip: 2
Strength: 1.4
Trigger Words
DcauGirls
Hash
AutoV2
037D32BAAD
default creator card background decoration
Second Birthday Badge
reevee's Avatar
reevee

LoRA for the Style used for DCAU Girls.

If you like my work and want to support me, please consider supporting me on my Ko-fi page. I'm open to donations and commissions for art and LoRAs with no censorship. This text is a link by the way :)

I trained it this with approximately 100 pics, which was a mix of screencaps and fan art. I trained it at 15 epochs with a dim/alpha of 32/16.

Not a requirement, but I would appreciate referencing me if you use my LoRA

This LoRA helps style females in the art style seen in the DC animated universe cartoons. The LoRA can work for older and younger females as the training data included both. The effect is somewhat subtle in certain cases, but more distinct at higher LoRA weights. Let me know how it goes for you. Read stuff below on guidelines and known issues :)

Triggers:

  • Input "DcauGirls" near the start of the prompt.

  • Input different colour hair, eyes, skin tone, hair style etc and you should be able to create original characters quite easily. Additionally, various unique superhero costumes are easy to create.

Prompting Tips:

  • Input "lips" in the negative field to make the characters look much younger if you want.

  • To get this to style a character LoRA you may have to reduce both LoRA weights significantly.

Notes:

  • My experience has shown scale 1 ~ 1.4 works best; my best results were 1.35.

  • If combining this with other character LoRAs consider reducing the weight of this LoRA to 1 or less.

  • My prompt order shown in the sample pictures worked best for me.

  • Works on different models though weights may need to change for good results.

  • I tested this on the AnyLora model and Midnight Maple model. Midnight Maple gave best results for this LoRA for me.

  • Refer to my pictures for examples and what upscaler I used, keep in mind I have used other textual inversions to help with poses and negatives.

  • I think this turned out well at least for the creation of original characters or DCAU characters. If there seems to be a high demand for a new version that more flexible then I will consider updating this.

I don't think people read entire descriptions like this, but congrats if you do