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 V20→V21
(This is not an exhaustive changelog)
General:
Improvements and Fixes
Improved Node/Group alignments
Added SAM Loader
Adjusting default values (e.g. nsfw detection model updated)
T2I:
Added Apply Signature
Added Epsilon Scaling
Basic:
Added VPred Patch
Added easy access for more advanced Detailer options (e.g. bbox crop factor)
Detailer:
Added easy access for more advanced Detailer options (e.g. bbox crop factor)
VPred:
Removed VPred Workflow (VPred support added to Basic)
FAQ
Comments (8)
continua sendo o melhor workflow que existe
Hi ! Do you think that you can add the node to change the aspect ratio of the image generated to the workflow ?
Like the SDXL Aspect Ratio node.
Have a great day !
Thank you for the suggestion, but I probably won't add the node. Some models already generate at higher base resolutions, so the node would obfuscate the process and benefit only certain checkpoints. It's also easy for someone to implement it themselves if they really want it.
Using Yolo 9, the pretrained/original SDXL 0.9 VAE, and ESRGAN (Remacri is) in 2025 is a joke, consider looking for better models for better results
what is better to use than remarci? that has been my go to. (and which for anime vs realistic if not remarci?)
and same for yolo models. i found yolo11 for face, but cant find newer hands/eyes
We would appreciate it if you have suggestions for what is better. So far so good with the suggested models in these workflows for me, but I'm always looking to try something new.
Thanks, man! this stuff is remarkable