Comment on Elon Musk’s Grok Goes Haywire, Boasts About Billionaire’s Pee-Drinking Skills and ‘Blowjob Prowess’

<- View Parent
brucethemoose@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

Since there is generated video, it seems like someone solved this problem.

Oh yes, it has come a LOONG way. Some projects to look at are:

github.com/ModelTC/LightX2V

github.com/deepbeepmeep/Wan2GP

And for images: github.com/nunchaku-tech/nunchaku


I dunno what card you have now, but hybrid CPU+GPU inference is the trend days.

As an example, I can run GLM 4.6, a 350B LLM, with measurably low quantization distortion on a 3090 + 128GB CPU RAM, at like 7 tokens/s.

You can easily run GLM Air on like a 3080 + system RAM, or even a lesser GPU. You just need the right software and quant.

source
Sort:hotnewtop