Remix.run Logo
bertili 5 hours ago

Is there any (near future) technology that would permit burning this terrabyte into some kind of ROM chip?

brute 4 hours ago | parent | next [-]

The path to ubiquitous AI (17k tokens/sec) https://news.ycombinator.com/item?id=47086181

You can still try it at https://chatjimmy.ai/, but it's running the rather outdated Llama 3.1 8B

antiloper 2 hours ago | parent | prev | next [-]

Given how fast models are improving, burning the weights into actual ROM is prohibitively expensive if you need (or want) to replace the chip every couple of months.

The alternative is on-TPU flash for storing the weights.

davidricodias 5 hours ago | parent | prev | next [-]

Yes, from 6 days ago: https://news.ycombinator.com/item?id=48986351

trollbridge an hour ago | parent | next [-]

A question of course would be "is 15,000 tok/s Gemini better than 100 tok/s Opus 5"?

quantumleaper an hour ago | parent | next [-]

It's better for the billions of free users that Google serves.

pimeys an hour ago | parent | prev [-]

Depends what you do. We have certain tasks we spend money on where Gemini 4.6 definitely is better than Opus 5.

bertili 4 hours ago | parent | prev [-]

That looks promising! As models become a commodity, this may turn out to be the real AI gold rush.

trollbridge an hour ago | parent | prev [-]

Yes. We're still a ways off from this being ubiquitous, but I am convinced it's coming.