Can you link to something so I can read more about this please?

Thanks for doing the legwork!

You spelt cunt incorrectly.

Is it only AOC being vocal?

Oh yes. The LLM will lie to you, confidently.

What are the local use cases? I’m running on a 3060ti but output is always inferior to the free tier of the various providers.

Can I justify an upgrade to a 4090 (or more)?

Great for turning complex into simple.

Bad for turning simple into complex.

Treat LLMs like a super knowledgeable, enthusiastic, arrogant, unimaginative intern.

What if we were able do to travel around in a robot suit that we could fully control. Oh yeah, thats a car.

But this is America

Maybe they hosted their servers in Eritrea, Turkmenistan or San Marino. No copyright laws there

Annas archive. Keep up. Pffff.

Agreed. Seed forever and release the AI weights and model. That would be fair payment.

The entirely of Annas archive would be an excellent benchmark training set. Particularly a cleaned processed dataset.

It’s easy to run a distilled version of the R1 model locally. It’s very difficult to run the full version. Min $6k to get 7 tokens per second.

Google was giving me bad search results about PTX so I just posted am opinion and hoped Cunningham’s Law would work.

Ah, I hoped it was cross platform, more like Opencl. Thinking about it, a lower level language would be more platform specific.

PTX also removes NVIDIA lock-in.

Not just deepseek, but everyone else who is now forking R1 and training for their specific use case.

There would have been many other runs before the release version.

Agreed. Privacy on the blockchain is an optional extra. On newer generations the option exists by design.

The data on the blockchain is not private.

However data can be encrypted before it hits the blockchain and it can also be cryptographicly manipulated in ways that remain private.