🛠️ 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
v7b workflow built and tested on:
ComfyUI 0.3.29
ComfyUI frontend v1.16.8
Python version: 3.12.8 (tags/v3.12.8:2dc476b, Dec 3 2024, 19:30:04) [MSC v.1942 64 bit (AMD64)]
Pytorch version: 2.6.0+cu126
Disclaimer: If you are using a different version than of any/all of those listed above, then this workflow may not work for you. I can't account for every difference, since we are all potentially using different versions of something.
v7b changes:
Workflow assembled from scratch.
No copy/paste or holding alt and clicking to copy nodes.
Perturbed Attention Guidance is bypassed by default. You can enable it by clicking on it and either press ctrl-B on your keyboard or click on the bypass icon.
Bypassing it does save time on generation speed.
Turning it on adds about 10 to 15 seconds on a 832 x 1216 image on a single KSampler on a RTX3060 12GB.
In terms of the entire workflow, you would be adding an estimated 5 to 8 seconds on the 2nd KSampler, each tile on USDU since they are set to 10 steps by default, and for each detected face/eyes/etc on the Detailer nodes.
PAG can help make the image look better, but is the extra time worth it to you for a possibility?
CFG++Sampler Select is toggled on by default. You can toggle it off by using the Boolean Switch directly above the node. Toggling it off will switch over to using the Sampler Selector (Image Saver) node.
Please ensure you adjust any other parameters as needed, such as CFG, Steps, etc.
There are a couple nodes above the Positive Prompt for token count. The node that shows the count is kind of buggy resulting in showing two lines of token counts when running more than one generation. If you don’t want to worry about this, you can just delete the two nodes and it will not affect the workflow.
I dropped all the little notes from the workflow. After rebuilding the workflow from scratch, they are the only thing not added compared to the old workflow.
If you are wanting more information on the various settings/parameters for each of the custom nodes, you can visit the project pages of each one.
FAQ
Comments (9)
Group nodes are being deprecated soon. If you have updated your comfy to 3.29 and the frontend was updated to anything past v0.15.x (which it probably was). Then nodes like Image Saver will be broken or will cause your comfyui to freak out.
From my understanding: Image Saver is considered a group node.
I was able to get Image Saver working again by using a command line argument (--front-end-version Comfy-Org/[email protected]) to load the older frontend version.
The Image Saver node looks a little janky now, but it works as long as you don't mess with the connections.
Doing some more testing today. I spoke with the author of the Image Saver node used in the workflow and (of course) after testing it with a super basic workflow, the node works fine.
Will do more testing to see what is causing the issue prior to the Image Saver node.
Testing done so far:
1. Updated comfyui and python dependencies.
2. Started with an empty workflow adding in all the standard Image Saver required nodes, a single KSampler, and 2 ImpactWilcardEncode nodes for the prompts since this is what Image Saver recommends for inputting the prompts into the Image Saver node.
3. Changed KSampler to SamplerCustomAdvanced
4. Added CFG++SamplerSelect
5. Added Boolean switch to toggle between CFG++SamplerSelect and Sampler Selector and Widget to String to feed the CFG++SamplerSelect sampler_name info to Image Saver which is also connected to the Boolean switch.
6. Added Image Comparer nodes.
7. Added 1 USDU node.
8. Added a second USDU node.
So far, no issues as of this message.
Workflow is basically fully rebuilt with minus the watermark portion of the flow and bookmarks. Still no issues.
Watermark Removal flow added. No issues.
All that's left are the bookmarks and the groups.
New version posted. Fully tested as noted on the description.
Testing out a workaround for anyone running into issues with the BasicScheduler node. Will update here if the workaround works.
Workaround appears to be a success. Will upload it soon.
Here's what was done:
1. Removed the Sampler Selector node.
2. Removed the individual Steps and Denoise nodes that were also feeding into each BasicScheduler node.
3. Moved the BasicScheduler nodes next to the nodes they are connected to. These nodes control the parameters for Scheduler, Steps, and Denoise.
4. Connected 3 Widget to String nodes to the initial BasicScheduler node to output to the Image Saver node.
5. 2 of the Widget to String nodes had to be connected to additional nodes in order to connect to the Image Saver inputs. 1 for String to INT and 1 for String to FLOAT.
6. Ran a test generation and all the info displays correctly when uploading to civitai.
7b2 has this adjustment made and a secondary JSON without ComfyUI-PPM.
If you were not experiencing issues with either, then 7b should still be fine to use.