Looks really fun, thanks for sharing!

Interesting. In my experiments, it seems to me WAN was trained on a lot of synthetic Human motion, rather than actual Human videos, and even with this LORA one can see the under-lying animation rigging that doesn't match how Human bodies really distort.

Any chance of a Hunyuan version please? 🥺

Is there a trick to using your LoRAs? I often get a rainbow pattern instead of a video.

Thanks! It's hard to find even 3d crawling animations online that are any good, so was hoping for this

I'd like to humbly request an i2v version. not to come off as ungrateful for your work but i haven't had a lot of luck doing i2v unless i'm missing something... great concept though, much love for making this

this is a really neat concept.

I2v version, immediately. Don't make me beg you T-T

Would like to thank you again for your work, and thank you for your strict and rigorous naming convention, when I look into my wan folder I know instantly which LoRAs are yours ^^'. THANK. YOU. <3

Banger after banger Playtime does not miss like prime Kobe!