Well, this is trained on a specific poses and close-ups. It is trained at maximum resolution, the quality is very good. There should be quite a few images where the futanari has cum on her chest and facials, it should be able to do it by saying something like " white viscous fluid on woman chest/breasts/face ". It is trained on close-ups, to make sure you see the face, describe it or put an action like " looking at camera ". It is not trained on typical keywords like futanari or missionary anal, you have to prompt it in a more descriptive way. According to samples it works very well, lets see what it can do!

Prompt example (add place and add woman’s attributes): The image is a high-resolution photograph depicting a sexual encounter between two individuals, a man and a woman. She is lying on her back on a bed with her legs spread wide apart, exposing her erect penis. She appears to be in a sexual act with a man, his erect penis penetrating her.