Not so Simple (or is it?) Wan 2.1/2.2 Lightspeed I2V Workflow

Minor change:

Swapped the WanImagetoVideo node out for the WanFirstLastFrametoVideo node.

Leave the end_image input disconnected and it will function the same as the WanImagetoVideo node.
This is another function that I do not use and will not provide any support for. IMO there are better workflows that incorporate this function, but people keep asking me to add it.
Edit: I got it to work pretty consistently using the same image for start and end images using DEIS + Simple and keeping it at 5 seconds (81 length). It absolutely has the flash "glitch" if I go any higher on the length. I tried other sampler + scheduler combos without success. YMMV.
Seriously, just use darksidewalker's workflow instead if you need/want this function.

Other than that, this workflow is exactly the same as v8.

The model these workflows are meant to work with are the Lightspeed versions of Wan 2.1 and 2.2 by DarkSideWalker

First Steps:

Select your high and low models.
If they have the VAE and CLIP baked-in, make sure to toggle the CLIP and VAE switches accordingly.
The option to use loras is available via the Power Lora Loader.

Next:

If I want to make another video to combine into a longer video:

Combining videos:

Toggle off the Image to Video group.
Toggle on the Combine Video group.
Place video 1 in the first Load Video node and video 2 in the second Load Video node.
Adjust the settings as needed, but it should be good to go without adjustments unless you have some special format, etc., that you are going for.
Queue it up.
You now have a combined longer video.

Rinse and repeat as needed.

Q: Why don't I save/extract the Last Frame at the end of the Image to Video group?

A: I don't always need, want, or require it. Especially, if I am just doing a quick-short video.

Q: Why don't I do A, B, and C, like does?

Answers:

I'm just getting into WAN and Video generation in general.
If I am just going to copy someone else, why would I bother creating a workflow?
I enjoy the process of learning and discovering while spending hours trying different things that possibly already include(d) what you probably are wanting to suggest to me.
I want to keep my workflow simple.

If for some reason you are getting OOM:

You can try adding "Clean VRAM used" nodes from comfyui-easy-use or whatever you prefer for that function.
Those weren't helpful for me, so I will not be including them.

v10 contains nodes from these: