Hacker News Reader

I've ran hunyuanvideoai from their GitHub and it seems to generate realistic videos. It is a bit slow (30-60min per video clip) and requires ~50GB VRAM. I wonder how the quality compares though?

View on HN

oh, and the output videos that are generated are 5sec clips at 554x960px. this is on a single A6000

It looks good, I'm just wondering why it has no attention from the ML community

HunYuan is all everybody on Banodoco and the broader comfy ecosystem are talking about. And that's with Lightricks' LTX model having just been announced too.

HunYuan is seriously amazing and it looks like it'll be the Flux/Stable Diffusion of AI video.

Sora is cooked.

It depends on where you're looking haha.

Reply on news.ycombinator.com