Download
1 variant available
This is an experiment at style type LoRA.
Update: This style is notoriously hard to make a model of. All its nuances and intricacies get lost in the training and the result is a very subdued version. Illustrious version worked quite well.
Will try Noob vpred and upload that too.
I've trained it on a friend's artworks + some images generated through Midjourney.
I am not too happy with the results, because it needs a more focused database, and probably higher number of steps. Think of it as an alpha version.