Z-image sfw to nsf.w controlnet inpainting

hey guys, i have this z-image inpainting workflow with controlnet and it works somehow decent, but especially for nsf.w it doesn't reliable produce good quality.

I am trying to create a male model by using sfw images and inpaint them.
Any idea on how to improve this workflow, or do you have one with inpainting + controlnet that is good (doesn't have to be z-image necessarily)?
thanks

https://redd.it/1s49lzt
@rStableDiffusion

Another interesting application of Klein 9b Edit mode

https://redd.it/1s6kjmv
@rStableDiffusion

SEEDVR2 - The 3B model :)

https://redd.it/1s6k735
@rStableDiffusion

PixelSmile - A Qwen-Image-Edit lora for fine grained expression control . model on Huggingface.

https://redd.it/1s62g0z
@rStableDiffusion

StabooruJeffrey SJ26 Q1: Quick Recap
https://redd.it/1s6rknr
@rStableDiffusion

I created a node to blend multiple images in a perfect composition, user can control the size and placement of each image. Works on edit models like Flux Klein 9b.

https://redd.it/1s6cvd6
@rStableDiffusion

Need help - transitioning from ChatGPT image Gen to SD
https://redd.it/1s6r6rh
@rStableDiffusion

LTX 2.3 Reasoning VBVR Lora comparison on facial expressions

https://redd.it/1s6uthp
@rStableDiffusion

Hunger of "Workflow!?"
https://redd.it/1s6uh4t
@rStableDiffusion