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 V18→V19
(This is not an exhaustive changelog)
General:
Improvements and Fixes (e.g. BodyADetailer Bug)
Adjusting default values (e.g. increased base resolution)
Changed face detection model (face_yolov8m→face_yolov9c)
T2I:
Added Simple Background Remover
Added easy access for more advanced Detailer options (cycles, max size)
Resize Regional Image before processing to avoid errors
Detailer:
Added Watermark Remover
Added Inpaint and Outpaint
FAQ
Comments (8)
Bro, this workflow is super complete, Many thanks. However, I’d like to ask if it’s possible to use or adapt the LoRA Loader (LoRA Manager) in this workflow? It’s much more visual for people who create LoRA recipes and don’t want to scroll through an endless list of LoRAs XD.
Usually, I make these connections myself in simpler workflows, but this one has so many wires that I’m honestly afraid of disconnecting something and then forgetting where it goes XD.
Anyway, thank you for the workflow, it’s wonderful.
The best anime workflow i've found. Keep up the great work!
AttributeError: 'NoneType' object has no attribute 'model_options'
AttributeError 291 FaceDetailerPipe
How can i fix that?
I changed my OS, and had to install this workflow again. And I get the same error as you. It seems there is an updated node that breaks the detailers. If I don't use any detailer the workflow works for me
Could you please provide more details, such as the settings you were using?
Just fixed it by deleting then reinstalling from git the custom node impact pack, then in custom_nodes/ComfyUI-Impact-Pack/modules/impact/impact/impact_pack.py
line 304:
model = nodes_differential_diffusion.DifferentialDiffusion().apply(model)[0]
changed it to:
result = nodes_differential_diffusion.DifferentialDiffusion.execute(model, 1.0)
model = result.model if hasattr(result, 'model') else result[0]
Edit: fix was from a the github impact pack page /ltdrdata/ComfyUI-Impact-Pack/issues/1113 (sorry idk if I can post links)
and I got to credit thomas-lennon-atu-ie for the fix :)