All the following images have been generated using the site: ( Ver. 2.1.3 )

Smile and suit are translated nicely. The hard shadows in the photo are translated into the hair by the ai.

Unfortunately, the AI automatically detects the face and crops the image (so no swan). It handled the b&w image quite well.

Earrings and hairline is pretty spot on. The teeths are not translating at all and the lipstick has moved down to the chin.

Double face. Happened to me a couple of times.

Another double face. The face behind actually looks to be going quite well with the beard, hair and skin color.

The highlight on her face gets very light. But besides that, the AI translates a lot of the expression really well.

AI gets the dark/white hair across along with lips. (AI had some trouble with red lips)

It did detect the hair despite being green and black. The hair was also was translated quite nicely to the output. However, the style of the output have less hard stokes and is more like water color.

Don’t know what is going on with the green drops. But the AI has done a great job with the shadow on the side of the head. It also nails the hair and beard. The eyes are really well translated to resemble the photo and achieving the anime/manga style.