I organized this workflow to help myself understand the example published on GitHub for the "ComfyUI-Wan22FMLF" custom node.

Using "ComfyUI-Wan22FMLF" and other custom nodes, this workflow allows you to perform Image-to-Video (I2V) generation by specifying 3, 4, or 5 images as keyframes within the video.

You can download the official workflow here.

https://github.com/wallen0322/ComfyUI-Wan22FMLF/

I do not yet fully understand the details of Wan22FMLF myself, so I may not be able to answer your questions. Please refer to the author's GitHub or other resources for more information.