LoRA training cost Civitai Buzz. If you like my work, please consider tipping me so I can keep creating them. Reacting to my images, posting your images and using Civitai generator with my models can also help.
The model is designed to generate high-quality illustrations of female or male characters, the style take inspirations from illustration of "Oshi no Ko" It excels at creating detailed character designs with simple backgrounds.
Issue : The Pony version can sometimes create kind of long neck, I don’t really know why.
Update: new version with FLUX 1.D with a lot more accurate style.
The V1 of this model doesn't know the Oshi no Ko characters and is not really good with hands.
Update 2 : The V1.1 is out. It can sometimes be more accurate with OnK characters (included in captioning). But it may also be a bit less flexible so do your tests with V1 and V1.1!
Starting the prompt with "a Mengo Yokoyari style image of" do work pretty well for me.
Sometimes, simplier is the prompt, better is the result.
Trigger words : Mengo Yokoyari
Description
V1 of a Flux 1.D version trained on Oshi no Ko illustration.
Is pretty bad at hands for now, will try to fix that later.