Makes colorful drawing-like images similar in style to the artist Witchparfait.

How to use
Prompt as normal. The dataset contained a fairly large amount of 1girls in various sizes and outfits plus SFW and NSFW so most related tags should work fine. There was some theme variety mixing sketches, greyscale and color but I believe I managed to tag everything to maintain style coherency.
Worth noting is that the "sketch" should be reasonably well defined for those who want a rougher look.

For further prompt tips please see the sample images.

MISC
Trained on Pony XL with a dataset of 98 images at 1024 resolution.
All sample images are raw txt2img with no helper embeddings/negatives used. E.g. - things should look better if you do use them yourself.
The training was done on a small subset of art, as such the output of this LoRA is not an accurate representation neither in quality or style of the source material. Go support the artist if you want the good stuff.