What's this LoRA
This is a LoRA trained on a single image, increasing the weight tend to make change the style and adding an armor to the subject. This is part of my "shot" series of lightweight LoRAs trained on single images. You need to lower the weight, otherwise all the images will look similar (check the comparison picture).
All the images in the examples have been created using my custom model (https://civarchive.com/models/24149) and this LoRA only. All the examples have been created using the hires-fix or by creating the base 512x768 image, upscaling it to 1.5 and the applying the same prompt again as img2img
All the pictures have been created using a weight between 0.6-1.0.
The LoRA works also on NSFW (check last examples).
Training Picture
This is the picture on which the LoRA is trained (also AI generated):
Guides
Learn how to create pictures like mine with my step-by-step tutorials:
Support
If you want to support my work, you can check out my music here:
https://inzaniak.bandcamp.com/
You can also check my AI Art accounts, where you'll find daily pictures made with my models (and previews of new ones coming):
Description
FAQ
Comments (5)
I'm unclear what this lora actually does. Is it for the armor, look, color, pose?
Yeh, or, like, what is the single image you used ?
I've added the training image to the description. A LoRA trained on a single image tends to copy elements from that particular image, so if you use high weights it will change the colors, style and composition of the image to match the original one.
Interesting idea...so would it be possible to train something like this with a very revealing outfit and then use weight to determine how revealing you want the outfit to be?
Would be very interested to learn how to train LORA on one image.