The absolute state of AI results.

[BradZax]

Originally Posted by Botten [You must be logged in to view images. Log in or Register.]

after you labelled references to videos, articles and government notices as 'AI slop' you then believe there is in fact a shortage of helium problem. [You must be logged in to view images. Log in or Register.]

No i dont care at all Im asking you why you're posting all this crap about it.

[BradZax]

Originally Posted by Ekco [You must be logged in to view images. Log in or Register.]

yeah it(DGX Spark) runs big local models well due to having a shit load of vram, but less CUDA cores than a high end gaming GPU and the high end gaming GPU doesn't have enough ram on the card to run big models, so idk where the sweet spot is going to land on all this shit yet for like my use case with Kaia

i'm still looking into used server-grade gpu cards with a pci adapter and 3d printed cooling shroud to work in a desktop

so can we string some laptops together in the garage to make droids w/ it?

[You must be logged in to view images. Log in or Register.]

[Botten]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

No i dont care at all Im asking you why you're posting all this crap about it.

Information that happens to anger you is still information.

And my conference call is up in 5 minutes.

[BradZax]

It doesn't anger me it's you that angers me.

There is your problem again, you always think its someone else, the billionaires, information, it's you!

[Ekco]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

so can we string some laptops together in the garage to make droids w/ it?

[You must be logged in to view images. Log in or Register.]

the MacBooks are pretty good i hear

Yes, you can cluster ("daisy chain") multiple Apple Silicon MacBooks together to run massive Large Language Models (LLMs) by pooling their unified memory. While traditional daisy-chaining usually refers to video displays, in the context of AI inference, this means creating a distributed cluster where a massive model is partitioned and calculated across multiple Mac computers simultaneously.

[Duik]

the MacBooks are pretty good i hear

Yes, the Scottish computer industial complex shines above
all others.