This was a test on how z-image would react to a slider LoRA trained with the prompts "a person with dominant personality" and "a person with submissive personality".
I didn't add anything else, and the database consists of a set of 15 images generated with z-image with the prompt "person".
The effect is not very powerful, but we can still see the changes in the body language.
The samples weights range from -2.0 (reserved) to 3.0 (assertive).
I was a bit out of ideas on how to test this effectively so please let me know if you find it useful.
I don't think I'll do other versions of this, this was mostly done to understand how z-image perceives submissive and dominant personalities.