Atomix Wan I2V Workflow
A Simple Wan I2V workflow
Description
FAQ
Comments (7)
CLIPVisionLoader
'NoneType' object is not callable
try these model : Comfy-Org/Wan_2.1_ComfyUI_repackaged at main
Is there any way to control timeline in I2V Wan? i mean prompt scheduling? if yes, how? frustating the the AI must do all the storytelling
To my knowledge, Wan has not yet been able to control timeline.
Man, I was more excited to use this workflow than any other I've seen, you had everything, 3 loras, both model types, but this fucking ksampler is taking longer than any other workflow I've used and I still turned everything way down to even make it work at all. I usually generate 720p video in 10 minutes at 30 steps, 5cfg, and 97 frames, but its been 20 minutes at 15steps 4cfg and 33 frames and the Ksampler is only half way done and there's still more to do. I also don't think your teacache is set up right because theres an empty model node on it and my console isn't loading it. So idk how to fix it but if you can let me know I'd really appreciate it cuz I reallly wanna use this workflow. thanks man.
This is a simple workflow that only uses ComfyUI's basic nodes. If you want to use advanced features, you should try using the ComfyUI-WanVideoWrapper node.
This is the only workflow that really works for me. THANK YOU!