🛠️ Purpose & Design Philosophy
This workflow is a high-fidelity environment built for Illustrious. It prioritizes stability and professional texture over generation speed. It follows an "all-in-one" philosophy: configure your prompts, hit queue, and let the workflow handle the multi-stage refinement from start to finish.
Not for Speed: This is a heavy-duty refinement tool. If you want 2-second previews, use a basic workflow.
Personal Use: Built for my specific production needs. It is shared as-is for those who want a "set-and-forget" pipeline for Illustrious. Adjust the settings and models to fit your needs. The defaults on v19+ will be what I used for the sample images.
All-in-One Logic: The workflow handles generation, detailing, and upscaling in one continuous pass.
If you're needing something with more features: Try checking out Silly All-in-one, Multiple Characters With Regional Loras which is on civitai. I have no plans to implement additional features.
⚠️ Disclaimer & Compatibility
Install at Your Own Risk: Custom nodes can break your environment. I am not responsible for troubleshooting your specific installation.
ComfyUI: Built and tested on the non-app version. Desktop app users may face additional hurdles.
The "Your Version" Factor: Your node versions and environment are 99.9% likely to differ from mine.
Nodes 2.0: I do not recommend using Nodes 2.0. It creates unpredictable UI behavior; I will not provide support for issues involving this feature.
🤝 Support & Boundaries
I have not and never will monetize my workflow. CivitAI is the only platform I am maintaining this workflow on. If you find a version of my workflow posted and/or monetized anywhere and use it, then whoever posted it is responsible for providing you any support. This is the main reason why I don't monetize this: I don't want to be or feel obligated to provide support.
No DMs: DMs are disabled due to repeat spam. Please check the Discussions tab below; most questions have already been answered.
Modifications: You are free to hack this workflow apart. However, you are responsible for fixing it if it breaks.
Custom Requests: I do not make private workflows. If you need a custom solution, post a Bounty on CivitAI. There are many talented creators ready to help you for a fee.
Description
Added a very basic ControlNet & IMG2IMG group (same as 13e)
If you don't know how to use ControlNet or IMG2IMG, there's plenty of info out there.
ControlNet:
I basically don't use it and when I do, it's just as a more complicated version of IMG2IMG.
TL;DR for Image 2 Image in this workflow:
Put your source image into the Load Image node.
Adjust the denoise to something below 1.0.
Change the toggle on the Latent Switch to 2.
Click Run.
As previously mentioned:Â
The Global Seed and Global Sampler nodes will override any of the settings related to them across the workflow.Â
Using a sampler or scheduler not compatible with any of the nodes will result in an error.
You can bypass or delete these two nodes if you want to do these settings manually on each node.
CFG is tied to the “CFG” node to the left of the KSampler.
Delete this node if you want to set CFG manually on each node.
The detailers can be used for whatever you want as long as you have the detections models.
The detailer nodes used in this workflow take all detected items and do them at the same time.
You can use whatever detection model you want. You don't have to use what I use.
If you’re using the Nodes 2.0 feature: you are on your own.
Custom nodes used in this workflow:
FAQ
Comments (11)
Hello! Thanks for the update ^^
I just jumped from v13d to v15, and I can confirm it feels around 40–50% faster, which is awesome :D
I only have one question about the img2img part. I think I might be doing something wrong, or maybe I don’t fully understand how it’s supposed to be used.
If I set the denoise too high, I get good generations, but of course I lose the original image as a reference.
However, if I lower the denoise too much, the result ends up being a fully grey image :x
Am I missing a step, or is there a specific way img2img should be used in this workflow?
I'll try one in a bit and get back to you. Just waking up.
One thing I do see off the bat that I missed was a switch for the width and height, so you're going to have to select that manually still for USDU's tile size and I missed connecting the VAE to the IMG2IMG part. I'll update this shortly. Just wanting to test what you were encountering.
@Zephir0ss it seems to work fine when I use 0.5 denoise, while keeping a resemblance to the source image. I used the same settings I normally.
I'll put out another update soon fixing the missing VAE connection.
I forgot to ask, but are you prompting anything? That could play into it. I usually still prompt something when doing IMG2IMG.
Tested it some more with lower denoise (down to 0.15) and no greyness. You can try turning off Epsilon Scaling or using different settings in general if you're still encountering this (and add prompts if you aren't already).
@Gladas Thanks for the follow-up!
I figured out what was wrong on my side, and it lines up perfectly with what you mentioned.
The issue was that the VAE wasn’t actually connected to the IMG2IMG path.
I had created a separate VAEEncode node by mistake, but its vae input wasn’t connected to the main VAE coming from the checkpoint / VAE switch. Because of that, at low denoise values the latent had nothing valid to work with, which resulted in the grey output.
Once I reconnected the existing VAE from the workflow to the VAEEncode used for IMG2IMG, everything started behaving normally.
Now with denoise around ~0.2–0.3, I get a proper image that keeps the source structure instead of a grey/noise output.
So all good now — thanks again for pointing out the missing VAE connection, that was exactly it! :D
Glad to hear it. As I was about to add the missing height/width switch, I remembered that I had added in a better solution to not need that any more, so you shouldn't need to set the resolution manually for img2img. Basically, the width and height are detected before it goes into USDU and adjusts the tile size to match the original image size.
On 15c, the VAE is not connected in IMG2IMG. You'll have to connect it yourself if your copy of the workflow is throwing an error there.
I'll be replacing the uploaded workflow file shortly with this fix.
Fixed 15c file uploaded and replaced old file.
This just adds the missing VAE connection in the IMG2IMG group.
Thanks again G, I love your work fam <3