I completely redid the last version and have seen no issues with generating on a variety of models, both realistic and semi-real. If you're still having issues, please let me know. I apologize for the difficulty.
She left us too soon...indeed she's left the industry and is making a good living as a DJ and Producer. From oral to aural - nice!
If you like my work, I'd appreciate it if you could Buy me a beer!
If you are interested in my creative process, you can read more in my blog post HERE.
And if you're feeling particularly curious, I can create a custom Textual Inversion or LoRA just for you!
[ SDv1.4, 4e-03, 5100 steps, v/s .210 ]