Intro

This workflow uses similar concepts to my iterative, with multi-model image generation consistent with the official approach for SDXL 0.9. Basically, it starts generating the image with the Base model and finishes it off with the Refiner model. In addition to that, I have included two different upscaling methods, Ultimate SD Upscaling and Hires. Fix. Both have their advantages and disadvantages, although I find the Ultimate SD Upscaling to be more consistent.

This workflow is fully configurable, although it can be used right out of the box without making any changes to the settings.

What does it do?

• Generate SDXL 0.9 model images consistent with the official approach (to the best of our knowledge)

• Ultimate SD Upscaling

• Hires. Fix

• Fully configurable

• Toggleable global seed usage or separate seeds for upscaling

• "Lagging refinement" aka start the Refiner model X% steps earlier than the Base model ended

• Compare the outputs to find what works best for you!

• Quickly iterate on settings

Note: I have not found the "best" values for denoising strength, samplers, base model %, etc. Feel free to try changing these values and seeing what effect it has!

How to install it?

1. Download the workflow zip file

2. Extract the zip file

3. Launch (or relaunch) ComfyUI

4. Load the workflow by pressing the Load button and selecting the extracted workflow json file

5. Select Queue Prompt to generate an image

Note: If you do not already have the ComfyUI Manager extension installed, you will need to do this first. It will resolve any missing dependencies you have for you. See below for more info.

Quality of life add-ons

I highly suggest downloading this ComfyUI Manger extension as well for ComfyUI as it will make downloading other extensions MUCH easier moving forwards. In addition to that, if there are any dependencies that you do not already have fulfilled, it can auto-download those for you.

Download URL: https://civarchive.com/models/71980/comfyui-manager