Today we will use ComfyUI to upscale stable diffusion images to any resolution we want, and even add details along the way using an iterative workflow! This is the first of 2 workflows for upscaling that I tend to use, and this is the more flexible of the two. We will be using a custom node pack called "Impact" and it has a ton of nodes in it that are useful, so get used to seeing this one! We will also use the manager node, which we installed in a previous video (note I used "git fetch" versus "git clone", so keep that in mind) I have linked that below as well. We will use these to scale up images we love while adding details and even injecting additional prompts along the way!

This video also goes into some depth on using the provider node and pipes, but of which are quite useful but have some confusing parameters like pk_hook and so on.

If you are confused, check out the SDXL graph basics here:    • SDXL ComfyUI Stability Workflow  Wha...  

#stablediffusion #sdxl #comfyui #img2img

Grab some of the custom nodes from civit.ai: https://civitai.com/tag/comfyui

Here is the manager: https://civitai.com/models/71980/comf...
Here is the Impact Node Suite: https://github.com/ltdrdata/ComfyUII...

Grab the SDXL model from here (OFFICIAL): (bonus LoRA also here)
https://huggingface.co/stabilityai/st...

The refiner is also available here (OFFICIAL):
https://huggingface.co/stabilityai/st...

Additional VAE (only needed if you plan to not use the builtin version)
https://huggingface.co/stabilityai/sd...