The absolute state of AI results.

[Botten]

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

It's interesting to see that the anti AI arguments are coming from AI ITT.

It reminds me of that strategy of evil overlords like in The Boys when Homelander was giving compound V to terrorists to make super heros seem neccessary.

Like AI out here, "Also tell them helium is running out—and that Im going to use all the helium. That will really get you two angry at each other. And neither of you know fuck all about heluim so it will sound smart."

Only you don't need AI to tell you on this thread there is a helium problem from the already shared videos, articles and our government's notice.

Of course your ilk believes it is all fakenews if the facts make them look like lunatics. [You must be logged in to view images. Log in or Register.]

[Reiwa]

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

Only you don't need AI to tell you on this thread there is a helium problem from the already shared videos, articles and our government's notice.

Of course your ilk believes it is all fakenews if the facts make them look like lunatics. [You must be logged in to view images. Log in or Register.]

BLM was in charge of the stockpile under dept of interior, which is also USGS.

So it's just one commie subagency sticking up for another commie subagency.

[Botten]

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

BLM was in charge of the stockpile under dept of interior, which is also USGS.

So it's just one commie subagency sticking up for another commie subagency.

BLM IS part of the Department of the Interior.

And the USGS IS also part of the Department of the Interior.

But... purity requirements are the real, technical issue and that part has nothing to do with agencies.

A damaged MRI superconductor comes to mind and “stick up for each other,” that doesn’t magically erase its need for helium purity.

[OriginalContentGuy]

Your dad is a noble gas.

[BradZax]

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

[BradZax]

So what's the deal with this new Nvidia computer? My understanding is that it.... does a local LLM?

It uses its ow2n processor and can run offline? I am having a hard time understanding what they're selling.

[Ekco]

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

So what's the deal with this new Nvidia computer? My understanding is that it.... does a local LLM?

It uses its ow2n processor and can run offline? I am having a hard time understanding what they're selling.

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

[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.]

[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.

[BradZax]

Relevant:



The metaphor here is you can't even argue your anti AI argument without using AI and Im saying your AI argument is just made up bullshit by the AI to get you to call me Ilk.

You're using the Westworld mother brain to justify west world mother brains dystopian security system.