following my simplest WAN 2.2 workflow with loras (turbo and NSFW), i did very similar workflow for triple length videos. It also works perfectly with 16gb VRAM.
just as previous workflow, this should work on any comfyUI installation,
You will need to download all models, vae, clip and loras yourself unfortunately.
What's good abut the workflow?
it's very simple. Input first frame, select video height and write three prompts.
It will save first and first two clips if you want, in case the last part was not good enough.
Remember that WAN doesn't really do great with parts over 5seconds, and that the quality and consistency loss is very possible when doing pure I2V. your mileage might vary.
how/why set it up.
first two parts are set as a normal i2v clips. You want MAXIMUM 5seconds length if you want to change the details between first and last frames. if you go longer, the scene WILL start going back to frame 1.
for the second/third part it will use last frame as it's first. again, 5s if you want to avoid going back, depends on scene/prompt. 7sec in default settings.
it requires a little luck to get last frame without high movement, or the quality of second clip will drop.
use 4steps for low changes in the background and 6 if quality is a bit low.
have fun.
turbo (lightning) loras are here: https://huggingface.co/lightx2v/Wan2.2-Distill-Loras/tree/main
version might be different when You browse it.
Description
Evolution, still as simple.
now works really well for both long and short videos.