Note: I spend a lot of time on these to make sure they are top notch. If you have ANY issues, I'm very responsive. hit me up and I'll resolve them for you.
This is an advanced workflow. It is node heavy and designed for multiple options and to return professional high quality results. If you are looking for simpler workflows, they are here:
An essentials one can be found here: https://civarchive.com/models/2349309/illustriousponysdxl-essentials
A simple one can be found here:
A complete user and troubleshooting guide inclusing a list of common error messages can be found here https://civarchive.com/articles/24678/how-to-use-my-workflows-a-comprehensive-breakdown
Instagram: https://www.instagram.com/synth.studio.models/
Buy me a☕ https://ko-fi.com/lonecatone
This represents hundreds of hours of work that I give away for free. If you enjoy it, Please make sure to 👍like and post to my workflows. That is how you support me. Also, feel free to ⚡tip 😉
也有中文说明
VER 10.1:
Got tired of having to always mess with resolution, so I gave you the option to make it automatic
Math node replacements to deal with the Linux errors
Added psst... fast bypassers (you'll thank me).
💡use the upgraded face yolo I added. much more accurate.
Improved notes throughout
Ver 8.0/ 9.0:
Major overhaul to bring it up to speed with my other workflows
Added second checkpoint metadata
added a prompt to hi-res fix
added Inpainting suite from the pro grade workflow by request.
V7.0
Apparantly , I forgot to hook up Face Swap in my last model and nobody told me until today, so a free major upgrade it is!
Installed a sentence remover in the prompt assist area to help eliminate CivitAI 🚩
Cleaned up some bugs
Resinstalled Qwen VL Prompt Enhancer
Upgraded Face Swap
also upgraded notes on how to remove it.
Minor changes to detailers
Added an inline crop tool in the post production suite.
Ver 6.0:
Major overhaul to bring it up to speed with my other workflows
Added a refiner checkpoint that can easily be switched on or off
Added FaceID as well as a style and composition IP Adapter for character consistency
Reowked the Detailers ro improve them
better version of Hi Rez fix
Reworked the Save with Metadata to align with CivitAi and fix the Comfy update bug 🙄
This is an all in one workflow:
Controlnet
I2I with prompt generation
Dual Model Loader: Better control
Sage Attention: Speeds up production
Ultimate upscaler: For prescale and hi-res fix
Detailers: Face, eyes, hand, expression
NSFW Detailers: Male & Female anatomy
Seed VR2 Upscaler: Get up to 4k pics, even on low VRAM
Post production suite: My pride and joy. Does everything.
Description
modified prompt assist area
removed sentence remover and placed two tag filters before the metadata instead
Added the ability to have two checkpoints shwo on CivitAI metadata
Added a prompt area in hi-res fix. I noticed that it helps realism to prompt for skin
Upgraded face detectiom to a better version (update hyperlink in notes)
cleaned up the Post Processing suite.
FAQ
Comments (11)
Hi, this is a very cool workflow, you have done wonders. i was wondering what the difference is between this workflow and your other pro grade workflow here.
Is it just the inclusion of the regional prompter and exclusion of the nsfw detailers? also for some reason whenever i do install your workflows and the required nodes and node suites it seems to ruin mmaudio on my other workflows? Do you have any suggestions or advice for me to correct this or is it just that mmaudio is not compatable with a particular node thats required? Im not exactly a noob however im not the most advanced user either. This is extremely advanced for someone of my level but im sure with some tutelage i can get it running smoothly lol. Again well done it is a very impressive workflow, so much better than anything i have been able to make thus far.
This one has regional prompters as well as multiple faces. The NSFW is all about the NSFW detailers.
The MMAudio issue is a global thing. It requires an older version of numpy. It will crash on a lot of new nodes when it updates as that repo is not well maintained. ConfyUI broke it and my workflows are up to the latest version of everything when I release them.
You can downgrade numpy by force reinstalling a verion before 2.3. Feed your log into Grok or Claude and fix all other errors first. I have mmaudio running fine, but it was a little bit of work
just to clarify. This doesn't have all of teh artistic stuff, just the base and the detailers. My pro grade one has everything.
Btw, you can cut & Paste the entire suite and should need no modifications
Im not shy to a little bit of effort to get things working just so long as i can understand the instructions being given im usually ok here and there. Thank you for the advice but what do you mean by cut and pasting the entire suite, do you mean taking a section of one workflow and adding it to the other canvas or am i being stupid here. Sorry to be a pain asking questions as im sure your busy but i want to learn is all.
@Feral_Signal_Studios You can move or change teh order in any of my suites (specifc to the model type) by changing the image getnode to the suite image setnode you want before it.
In this case, all of the getnodes align with my base model, so at any time you can cut out an older version, or a version from that same model type and replace it.
⚠️ Make absolutely sure you remove the old one BEFORE pasting in the new one or it clones the setnodes.
I have this spelled out in my "how to use my workflows' guide. Unfortunately, I cannot link it as CivitAI is having issues, but it is on my profile
@lonecatone23 Thats pretty cool. Ill keep playing about with things to see if i can get it running. Ill give you another shout if i get stuck which will happen for def im sure hah. Thank you again for the help and advice, you are a star!
Maybe some setting details have changed, but I really need to confirm that, why after I update the workflow in version 8.1, the consumption of VRAM goes higher than before?
I do confirm that the SEEDVR2 settings were the same as version 7.
By the way, when I close the QWEN junc in version 7, the workflow performs better than before (really different than version 6).
Hope someone has a chat on this topic~ Your advice is more vital~lol
PS: My current laptop GPU is an RTX5060-8G laptop,.
Thank you very much ~~~~
Qwen issue: Xformers changed. It irritated me so muhc I pulled it out of my workflows for a while. Unfortunately, teher are zero other local options (Olama, Gemma, etc.) That are not significantly worse. I plan on messing with Gemma 4 and seeing if it is a viable replacement.
Where is it consuming more VRAM? I've made no significant front end changes
@lonecatone23
In seedvr2 with 3B+gguf module, the same settings in version 7 (RTX5060-8G laptop) when generate 1080*1035pix worked well, but in version 8 the comfyui announced to me that the VRAM overload, even I followed the note next to it to change and check.
It is so strange~ lol.
By the way, QWEN junc in 5060 still plays as a trouble-generator. Under the "prompt process", it changed my original mind. Sometimes I check to find that, there may be some security checks in it, especially in QWEN3.5 or higher. I have no idea where it comes.
@1648003234289 Yeah.... sigh.... 🙄
Put your log through an LLM to see what the issue is. I have zero issues, but I'm running with an RTX5090 and 128gb RAm, so it's really hard to diagnose vram issues
@lonecatone23
Thanks a lot, I will check later~ lol~