I trained this one to be flexible & play nicely with different styles. It works well with photos, anime, 3d, etc. However, the hairstyle requires some prompting and the face isn't 100% consistent. I'll have to do another later with more consistency.