NOTE:
I am uploading this so that someone smarter than me can improve on it and we can have a DANK SDXL animatediff workflow. If you do so, please share it with everyone as I did mine :)
This is the SDXL version of my AnimateDiff workflow that many people have come to know and love. It includes all the same goodies like a specific alpha mask for the subject so you can accurately use subject differentiating IP adapters. The groups are all labeled and it should be pretty easy to navigate.
A few considerations:
SDXL and HotShotXL have always been harder to work with than 1.5. This workflow specifically works with the SDXL 4 Step Lightning LoRA, so you will need to download that. Also, the only model that has really worked for me with this has been DynavisionXL. If you choose to use anything outside of that model and LoRA, that's your call.