Here's a workflow for SDXL designed for creating highly detailed images in all styles, but primarily focused on female anatomy.
It's the result of my long hours of practice on ComfyUI, applying two upscaling passes intermixed with two detailer passes for various body parts. I tried to keep it as simple as possible, but it's a complex workflow that isn't necessarily easy for a new user to grasp.
Here's a list of the different models needed to use it:
Detection models (Folder also present in the archive provided with the workflow, put them in /models/ultralytics/bbox)
Xinsir Tiled Controlnet (rename it as "Xinsir_controlnet-tile-sdxl-1.0.safetensors")
Lora for Face detailer you can use your character's Lora instead of this one.
All nodes used should be able to be downloaded via the Manager but I have not tried with a fresh install, I will provide details if necessary.
You can change the aspect ratio for images of various formats, and a landscape true/false button is used to calculate tile sizes for upscaling.
I recommend a base image resolution between 1 and 1.2 megapixels, depending on your preferences, the output image will be 3x larger.
You can also mute/bypass the Grain filter at the end of the chain if you wish.
Leave a review or tip me if you wish, it will be greatly appreciated, I give you here all my experience and my know-how on ComfyUi in order to obtain the best possible images with SDXL.
I hope you have fun with it :)
Description
Added I2I functionality and reorganized the detailer prompt nodes to make them easier to modify. Made some changes to the Lora used and tweaked some settings. Have fun ;)
FAQ
Comments (11)
make for wan video plz
On the way for Wan T2I
No Changelog? Can you tell us please whats changed in v2? Would be great. This is my main workflow at the moment for txt2img. :-)
in about this version section : "Added I2I functionality and reorganized the detailer prompt nodes to make them easier to modify. Made some changes to the Lora used and tweaked some settings. Have fun ;)"
@graam558 Sounds really great! Will try it out and give feedback. :)
hi , great workflow ! i want to make a lora character gen with this wf , the first ksampler preserve the details but once the tiled diff and control net occurs , i loose the original details . I also want to keep all the rest after ( detailers ) but keep my main original face . Any solution ?
Thanks for that great workflow!!!!
unbelievable workflow... any chance of having a complete dmd2 version of it? I tried to modify it, but uspcaling with dmd2 always ruins the picture...
Tiled diffusion is not compatible with DMD2; you would need to change the process to an ultimate upscale or something similar, and even then I'm not sure it would work, and you certainly wouldn't get the same results.
@graam558 thanks for the reply ;) i'll test a bit
I've tried quite a few 'upscaler/detailer' workflows and they're all kind of meh. This one is crazy good! Thanks for sharing your work!










