Workflow for XL/Illustrious/NoobAI Models
Any feedback would be appreciated.
See the Article for a Guide:
https://civarchive.com/articles/17339 (probably doesn't work rn)
https://civarchive.com/articles/17339
Key Features:
LoRA
TIPO Prompting
Detail Enhancers
ControlNet Upscale
IPAdapter
HiresFix
OpenPose
Wildcards
ControlNET
Color Match
FBCNN
Regional Prompting
Inpaint
Outpaint
Workflows included:
🟥Advanced: comprehensive text to image workflow
🟨Basic: more basic version of the Standard workflow
🟦Standard: more basic version of the Advanced workflow
🟩Detailer: detailing previously generated or existing images
Experiment and enjoy!
Description
Update from V34→V35
(This is not an exhaustive changelog)
General:
General improvements and bug fixes
Changed default values
Advanced:
Added Sam 3.1 Detailer
Standard:
Removed Body Detailer
Basic
Removed Body Detailer
Detailer:
Added Sam 3.1 Detailer
FAQ
Comments (21)
I think it's a brilliant workflow, many thanks! However, in the latest versions I don't seem to be able to apply a lora in the regional prompter. Is that true or am I missing something?
Just wanted to say thanks for so quickly updating after the Comfy Update, much appreciated!
Quick follow up though, been doing some gens, and the advanced workflow seems way slower now, not sure if you know what change(s) you made that would force this, or if it'smore with the ComfyUI update. Either way, what was taking me ~3-4 minute gens is now taking ~an hour. Any ideas why?
Just a little nitpick, V35 QwenVL has a text replacer node on top of the Checkpoint Loader.
Edit: got fixed, thx!
Thanks for pointing that out, I corrected it
Thanks for the workflow.
But when i use faceID, output become messed up . Is it because i use different checkpoint model? or i should use faceID together with other nodes?
The FaceID adapters were trained on the base SDXL and "conflict" with fine-tuned models because the adapters no longer align correctly. I'll probably remove FaceID in the next iteration since it's an outdated remnant from previous versions that I just happened to keep.
hey i have one problem.... comfy has an error message : Unknown pack (2)
Fast Groups Bypasser (rgthree)
Fast Groups Bypasser (rgthree) and the thingi check thing isnt idk working
Do you have the rgthree node pack installed?
The Lora loader is broken. It loaded 1 LORA and REFUSES to acknowledge any others. I did a clean install got it working but only sees ONE LORA.
Workflow seems great only problem is when I generate an image it resets my positive and negative prompts, is there a way to fix this
Great Workflow when it works, it causing a lot of lag for me when i uses(probably due to sheer amount of nodes and string), this lag cause the prompts to lag out, forcing me to disable any upscaler functions. Now my generation are extremely oversaturated with red highlight when going through the adetailers(expect for mask Detailer and AnyDetailer), and sometimes the color fixer does it fix the image or other time it does not. Basic T2IMG works fine, the lora mananger, and prompts all work.
Excuse me, could you tell me what version Comfy UI this workflow works on?
From my tests, both ComfyUI v0.22.0 and v0.21.0 worked (I haven't tested any other versions)
Hi, only getting Basic v35 to work on ComfyUI 0.22
The other workflows show me this error message when loading:
"# ComfyUI Error Report
## Error Details
- Node ID: N/A
- Node Type: N/A
- Exception Type: Loading aborted due to error reloading workflow data
- Exception Message: TypeError: inputWidget.value.replace is not a function
## Stack Trace
```
TypeError: inputWidget.value.replace is not a function
at DOMWidgetImpl.callback (http://127.0.0.1:8188/extensions/comfyui-lora-manager/lora_loader.js:166:51)
at DOMWidgetImpl.set (http://127.0.0.1:8188/assets/dialogService-DSBgqcNn.js:2:29899)
at ComfyNode.configure (http://127.0.0.1:8188/assets/api-D9vMMk51.js:2:148864)
at ComfyNode.configure (http://127.0.0.1:8188/assets/dialogService-DSBgqcNn.js:153:11721)
at LGraph.configure (http://127.0.0.1:8188/assets/api-D9vMMk51.js:2:394254)
at oi.configure (http://127.0.0.1:8188/assets/dialogService-DSBgqcNn.js:178:8398)
at LGraph.configure (http://127.0.0.1:8188/extensions/comfyui-custom-scripts/js/reroutePrimitive.js:14:29)
at ComfyApp.loadGraphData (http://127.0.0.1:8188/assets/dialogService-DSBgqcNn.js:178:15190)
at async app.loadGraphData (http://127.0.0.1:8188/extensions/cg-use-everywhere/use_everywhere.js:144:17)
at async ComfyApp.handleFile (http://127.0.0.1:8188/assets/dialogService-DSBgqcNn.js:178:21709)
```
## System Information
- ComfyUI Version: 0.22.0
- Arguments: F:\Extras\Stability Matrix\Packages\ComfyUI\main.py --preview-method auto --enable-manager --use-sage-attention
- OS: win32
- Python Version: 3.12.10 (main, May 30 2025, 05:39:07) [MSC v.1943 64 bit (AMD64)]
- Embedded Python: false
- PyTorch Version: 2.12.0+cu130
## Devices
- Name: cuda:0 NVIDIA GeForce RTX 5070 Ti : cudaMallocAsync"
@lanscrew Try rolling back to version ComfyUI v0.21.0. If that doesn't help, I recommend reinstalling Comfi itself, first saving or moving custom nodes, models, and lore. I can't recommend a definitive solution. ComfyUI has been performing poorly for me lately.
@Yakooo Thanks, I'll try that and report back.
Hi! Sorry if this has been addressed before (couldnt find any mention of it). When loading all workflows besides Basic v35 on ComfyUI 0.22.0 the process gets aborted and I get the following message on a popup screen:
"Loading aborted due to error reloading workflow data
TypeError: inputWidget.value.replace is not a function"
Is it something on my end or something caused by an update? Any input would be appreciated.
I have since updated Lora Manager but now the error message has changed to:
|"TypeError: Cannot read properties of undefined (reading '0') ## Stack Trace"
Lora loader is broken
it seems the QWEN enhancer doesnt take the original prompt or am I missing something?