yes, my style

Graphics tablets drawing, high contrast, dynamic, chaotic, sketchy and messy.

Limited dataset. Mainly close-up portrait. Does not generalize well on wide images.

It was overfitted, thus cover images used 0.3x RDBT model to fix small errors.

Main trigger word: @reakaakasky

Potential trigger words: sketch, rough outlines, anime drawing, dynamic lighting, expressive brushwork.

Prefer simple, short prompts, less than 20 words.

This is my style. There are many like it, but this one is mine.

You can't avoid the inevitable arrival of AI, just as you can't avoid the replacement of paper and pen by graphics tablets. AI is another tool, just more difficult to use.

Looks somehow familiar? It sneaked in the RDBT model for several versions.

This style might get a ckpt version released in the future, if I can draw the face and eyes even better...