That would be nice, but it's asking a bit too much from poor old SD. It has enough trouble keeping track of the arms and legs of ONE human, trying to make TWO humans that interact physically almost always results in scary abominations. It has a chance of working if they just stand straight up next to each other and don't touch, but that's about it.

Glad you like it! I'm totally not the right person to ask what these models are good for though. Because I don't prompt normally like everyone else. I just like to throw one to three random words at the model that don't really tell it much, and let it be creative and come up with something on its own. Then I expect it to do a good job at it! And models often get a little wishy-washy about details on accidental things they added without you explicitly asking for them, so this is something I care a lot about. v10 is better at putting the final touch on such accidental content, for example more often getting clothing details correct. But it is less creative and wild IMO (and more heavily girl biased), and sometimes has a more grainy picture quality on things like foliage, and also fabric textures can look "cheap". But that's probably also just because I give it such shitty prompts to work with, and only give it 12 steps to figure it out. In the images posted by evtqyn912 I don't see any such problems, so it seems a more developed prompt and more steps fixes it.

So I just don't know how it will work for you, try them and find out which suits your taste best!

And don't forget to post your results here! ;-)