This is using Stable Video Diffusion 1.1 via ComfyUI.
I largely only use the stuff I can run locally, so it's not as nice as what some of the websites can generate but I'm also freed from any censorship or monetary restrictions.
Super easy, but the results are far more often glitchy miss than hit.
If you’re just doing it for a fun experiment go for it, if you’re wanting results maybe wait another 6 months to a year for some new models or methodology to come out.