🗒️ My Illustrious Workflow for ComfyUI
Hi, so I thought I’d share the Illustrious Workflow I just finished building.
This is an evolution of a workflow I’ve been using for a few weeks now.
Main Points of Function
Highres Fix: Set to 50% by default, but feel free to play with what works for you.
4 Detailers: Face, Eyes, Hands, NSFW. Each can be toggled on/off using the Yellow Fast Groups Bypasser node.
Metadata Saving: Images save with full metadata for CivitAI, including prompts, checkpoint, LoRAs, etc.
Disclaimer
I’m no expert, in fact, i'm an absolute novice. This workflow came from looking at other great ones by DigitalPastel and Mittoshuras, then I linked things together until they worked.
Workflows I built from:
⚠️ That means if you have trouble running this workflow, I probably won’t be able to support you as much as I’d like to, or even at all really.
I’ll write a short a step-by-step guide below, but if you are running into issues, ChatGPT is your friend (that’s what I did) copy and paste your error into that, and stick with it.
Alternatively, if you look at this workflow and think “what the hell is this” I seriously suggest DigitalPastels Smooth-workflow, It was an absolutely great start for me.
How It Works
Blue Image Saver
Set your save path.
If you end the path with
%date, it creates a folder per day to keep things organized.
Green Positive Prompt
The Positive Node is divided into 2 areas.
Top box = your positive prompt.
Use "Select to add LoRA" to insert LoRAs.
Format examples:
embedding:SmoothNoob_Quality<lora:Smooth_Booster:0.5>
Bottom box = the executed prompt (and Wildcards if used).
Red Negative Prompt
Same formatting as positive including embeddings.
Blue Input Parameters
Set STEPS, CFG, Sampler, Scheduler, SEED.
Blue HiresFix
Choose rescale method:
by percentage
set Width/Height
to longer side (maintain aspect)
Yellow Fast Groups Bypasser
Toggle detailers easily if you don’t want to use one.
Yellow Detailers
Separate Face, Eye, Hand and NSFW detailers.
Defaults work well, but tweak if you like. That said, have a play around, just make sure you remember what you change so you can go back if it doesn’t work.
Image Comparer
View final image vs. no‑detailer version interactively.
Helpful Models
Eye: https://civarchive.com/models/178518/eyeful-or-robust-eye-detection-for-adetailer-comfyui
Hand: https://civarchive.com/models/329458/hand-detailersegmentation-adetailer
NSFW: https://civarchive.com/models/1313556/anime-nsfw-detection-adetailer-all-in-one
Final Notes
And, well, that’s it. Other than to say, if you are having trouble, I suggest checking out Smooth Workflow as this was mainly built from that.
I hope this helps someone. I’ve really enjoyed making AI images these last few months, I always try to share as much information about my generations as I can, so that other can remake and remix them to their hearts content, and I’m hoping this will help with that as well.
Happy generating! 🚀
Description
FAQ
Comments (8)
Hey, what Lora or prompt you use for that sketch girl output?
none at all, that's what the checkpoint and workflow produced with a blank prompt.
Is there a denoise for hires itself? Or does the percentage take care of that along with the steps?
Also, is there a way to make the adetailer have seperate steps? I appreciate the response in advance
Not sure if this is going to be any good to you now, but the new version being posted today, v1.1, has it so that the detailers have separate steps.
TOP
I'm getting error when i try to run.
The error:
Prompt execution failed
Prompt outputs failed validation: Exception when validating node: tuple index out of range KSampler: - Exception when validating inner node: tuple index out of range Image Saver: - Exception when validating node: tuple index out of range
I did run through google gemini, it said that wire(the connection between modules) is bad, but when I tried to reconnect everything as before. When i did, some connection couldn't be reconnect. I was wondering if you have any help for this.
A new 1.1 version should be posted in an hour or so that should fix this.
@DungeonsAndDosages Appreciate it! Thank you.



















