This was one of the first working Lora ever trained. <- link is for the checkpoint
Lora version is part of DJZ Zenkai: https://civarchive.com/models/10003
It was published back on January 1st 2023, originally for SD2.1 & later retrained on SD1.5. I used cloneofsimo's repo, to train this (back when nobody knew or cared what Lora Models even were). He was the original Author of the first Lora Code, check my comments on his github. (Christmas Eve 2022!)
Now retrained in Glorious Flux.1
Full captions will be supplied in each version once they are collected into a text only format.
Find them in my Prompt Lists (DJZ-Nodes on Github) or check the "Training data" download option here on Civit.
Description
FAQ
Comments (2)
A fun LORA, even if does seem to have a 'ONE TRACK' mind. A vast collection of such LORAs, in all kinds of abstract styles, would be very useful for amateurs and pros alike.
Thanks - I'm reviving my 2022 art collection for retraining in Flux - the datasets are only small 10-40 images on average and feature a "tight spread" of these unique looks, so i like to teach people how they can take their own work and create this "estimation models" for use in blending elements by way of 3-4 lora stacks!
thanks again for the kind words :)
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.