Anthropic expands partnership with Google and Broadcom for next-gen compute

Post by franktankbank » Tue, Apr 07, 2026, 6:05 PM UTC

brokencode wrote: post #17

skybrian wrote: post #12

I guess gigawatts is how we roughly measure computing capacity at the datacenter scale? Also saw something similar here: > Costs and pricing are expressed per “token”, but the published data immediately seems to admit that this is a bad choice of unit because it costs a lot more to output a token than input one. It seems to me that the actual marginal quantity being produced and consumed is “processing power”, which…

Gigawatts seems like more a statement of the power supply and dissipation of the actual facility. I’m assuming you can cram more chips in there if you have more efficient chips to make use of spare capacity? Trying to measure the actual compute is a moving target since you’d be upgrading things over time, whereas the power aspects are probably more fixed by fire code, building size, and utilities.

Post by bdangubic » Tue, Apr 07, 2026, 7:25 PM UTC

zozbot234 wrote: post #122

MrOrelliOReilly wrote: post #114

Earlier quoted context omitted.

How so? Opus and Sonnet are frontier models which cannot easily be replicated. Compute has real physical constraints which require appropriate procurement at this scale. At least those two points seem like pretty strong moats against the majority of companies.

You don't need to "replicate" Opus and Sonnet, you just need to match their overall performance at lower cost. That's been absolutely doable so far, with a steadily decreasing lag time.

Post by MrOrelliOReilly » Wed, Apr 08, 2026, 8:28 AM UTC

zozbot234 wrote: post #122

MrOrelliOReilly wrote: post #114

Earlier quoted context omitted.

How so? Opus and Sonnet are frontier models which cannot easily be replicated. Compute has real physical constraints which require appropriate procurement at this scale. At least those two points seem like pretty strong moats against the majority of companies.

You don't need to "replicate" Opus and Sonnet, you just need to match their overall performance at lower cost. That's been absolutely doable so far, with a steadily decreasing lag time.

Post by yencabulator » Wed, Apr 08, 2026, 6:03 PM UTC

AlphaSite wrote: post #78

nomel wrote: post #26

Earlier quoted context omitted.

> but the published data immediately seems to admit that this is a bad choice of unit because it costs a lot more to output a token than input one And, that's silly, because API pricing is more expensive for output than input tokens, 5x so for Anthropic [1], and 6x so for OpenAI! [1] https://platform.claude.com/docs/en/about-claude/pricing [2] https://openai.com/api/pricing

I think for the same model wall time is probably a more intuitive metric; at the end of the day what you’re doing is renting GPU time slices. Large outputs dominate compute time so are more expensive. IMO input and output token counts are actually still a bad metric since they linearise non linear cost increases and I suspect we’ll see another change in the future where they bucket by context length. XL output contex…

Post by pepperoni_pizza » Thu, Apr 09, 2026, 8:40 PM UTC

delichon wrote: post #126

pepperoni_pizza wrote: post #104

Earlier quoted context omitted.

Gigawatts are units of power, gigawatthours are units of energy. The equivalent of cars would be pricing by how much gas you burned, not horsepower.

1 horsepower = 745.7 watts