Sign In

Verified:

SafeTensor

Type

LoRA

Stats

35

3

4

Reviews

Published

Apr 15, 2025

Base Model

SDXL 1.0

Training

Steps: 500
Epochs: 10

Usage Tips

Clip Skip: 1

Hash

AutoV2
052A149A7A
default creator card background decoration
Luminaria's Avatar

Luminaria

Created on Civitai

I was looking for a LoRA of Cad Bane, but for some reason, I couldn't get good results with the only one available on CivitAI. So, I decided to make my own. This is actually my first LoRA. I didn't have that many reference images to train it, so it's far from perfect, but I was able to get some decent results with it.

Originally, I didn't plan on sharing it, so it's not super versatile. It was mainly designed to work with the checkpoints I personally prefer (I mostly use Juggernaut XI and the Cheyenne models; I just love them!). I also tend to combine it with LoRAs created by the Cheyenne creator (like Scribouillard, Moutarde, etc. => https://civitai.com/user/Aurety/models), and some Eldritch-style LoRAs (comics, gold leaf, etc. => https://civitai.com/user/eldritchadam/models). They're all fantastic (huge credit to the creators!). So yes, this LoRA works best with "comics" styles. Hyperrealistic or photorealistic styles don"t seem to give good results, at least in my experience.

As I mentioned, this LoRA isn't the most flexible, you'll probably have trouble generating images of Cad Bane without his hat or coat, or with different facial expressions (like smiling). That said, I did manage to get interesting results from time to time, depending on the checkpoint or LoRA used. Feel free to experiment!