Stereotypen

Yes, that was very helpful.

Many thanks for the detailed procedure. I will see what results it leads to.

If everything fits, I will upload the result again.

Regards, Armin.

Hello @Art Directress,

Thank you for your message. I am sorry you are having this problem.

Your prompt is too long, too conversational, like a story. The model will not understand it well.  The most important object should be as far left as possible and everything related to that object should be grouped together.

I recommend this document: Writing effective prompts to further help you in crafting good prompts.

When I am constructing something, I start with what is the core or primary part of the image and build up from that. The more you add to a prompt, the more likely it will get the core off from what you want. So, at some point, you are left to decide whether it is easier to take what you have created and go to generative fill to add in other objects/background.

So, I started with what appears to be core in this case. I know nothing about this subject matter, so I could be way off, but just wanted to show starting with the core of the image and then working out from there. My basic rule is the main object is to the left of the prompt with qualifiers in front(usually adjectives) and back (usually prepositional phrases), and then followed by less important objects and background.

dark-skinned Australian aboriginal with legs formed together as snake tail 

Hope this helps.

My best,
    droopy