the breastexpansion subreddit is completely infested with AI expansion vids. Macromastiadreams on kemono doesn't exactly have expansion, but the gens are very high quality (for what it is). users darkroast and bebobular on Civtai have a good collection of gens, including models/lora. I'm still learning comfyUI and my pc sucks so my gens are very poor.

here's one I did today. Not terribly happy with it, but that's as much as I'm willing to fuss with it. original image from Freepik.

That looks great to me! Which AI software if best for this? Would like to give this a go myself!

>>38353 I'm currently finding some success with ComfyUI using the Wan2.2 image-to-video model. You can use your favorite screenshot, a generated image from a text-to-image model, or a carefully procured (read the license and liabilities) stock photo. You can also generate a video straight from text as well. You see stuff like this come from Kling, Sora, Grok and other non-local services, but that requires tedious workarounds sometimes. Expect to perform an unreasonable amount of trial-and-error as you deal with the unintuitive inanity of how the machine interprets prompts, I feel like it's a constant monkey paw situation.

Assorted imagery of an imaginary girlfriend inflating in the park.