I created the model myself from scratch to still have a slightly painterly feel without creating messy shading. I wanted to still maintain a clean separation between shapes despite the texture :)
I recommend keeping the settings at around 0.7-1.0
What's really interesting is that different models can give you quite different results, so I prepared a special showcase for this Lora so that everyone can use Lora exactly as they want!
The most potential and best for use with this Lora are models like:
(( Nova Orange XL v3.0, ))*
https://civitai.com/models/1194136 Illustrious SNAFU*
Red Lily | Illu*
Haruka v2 work too*
coco-Illustrious-NoobXL-Style (for more comic-like results)*
Otome v2 (more book illustration-like results)*
You can see examples in the showcase at the top or check out the collections under my profile
I programmed these words to trigger specific results: (Experiment!).
hard light,
complementary colors,
color contrast,
high contrast,
beautiful colors,
Hatching shading,
oil paint texture,
Flat color,
Pastel colors,
Neon Colors,
clean shading,
dark mood,
clean shading,
low contrast,
halftone,
sketch,
ink,
monotone,
Interaction between characters,
glitch effect,
dark background,
subsurface scattering,
glow effect.
To get rid of any potential errors and also get the best results, I suggest pasting this negative prompt:
modern, recent, old, oldest, cartoon, graphic, text, crayon, graphite, abstract, deformed, mutated, ugly, disfigured, long body, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, very displeasing, (worst quality, bad quality:1.2), bad anatomy, jpeg artifacts, signature, watermark, username, simple background, conjoined, ai-generated, mini_person, extra chibi
