Claude Sonnet 5 • Hacker News Board

Re: Claude Sonnet 5

#791

Post by boc » Wed, Jul 01, 2026, 5:05 PM UTC

throwaway27448 wrote: post #780

boc wrote: post #608

Earlier quoted context omitted.

> Anthropic has little to no chance of producing a competitive business model in the long term. Extraordinary thing to say about the fastest growing company in the history of capitalism. They will soon have access to public markets, essentially unlimited capital, and can build insanely large models that they don't have to make public... ever. They can just use those models to run their business, train better models,…

We both know that both revenue and especially valuation, especially for something with as little moat as software, can vanish overnight. This is a commodity market and anthropic is serving boutique software, and the market gets more crowded every day.

So you're arguing it's the Yogi Berra "Nobody goes there anymore, it's too crowded" of business models?

Re: Claude Sonnet 5

#792

Post by danenania » Wed, Jul 01, 2026, 5:18 PM UTC

rimliu wrote: post #745

danenania wrote: post #304

Earlier quoted context omitted.

I’d also point out that LLM inference revenue already totals more than 100B annually based on publicly reported numbers. Almost none of that is replacing knowledge workers. Almost all is increasing their productivity. So empirically what you describe is already happening to a nontrivial degree.

Where does this BS come from?

> publicly reported numbers

Anyone can google it /shrug

Re: Claude Sonnet 5

#793

Post by giancarlostoro » Wed, Jul 01, 2026, 5:41 PM UTC

maxloh wrote: post #761

giancarlostoro wrote: post #15

Earlier quoted context omitted.

The reality is that Fable will eventually be obsolete and Sonnet / Opus will surpass it. Fable did cost 2x as much as Opus, so I assume it involves a much higher cost for what it did, but I wouldn't be surprised if Fable will be obsoleted by Opus or even Sonnet sooner or later at less cost.

According to CursorBench [0], Fable is the first runner-up, scoring 72.9% ($18.02, Max), while Opus 4.7 Max hits 64.8% ($11.02) and GPT-5.5 Extra High sits at 64.3% ($4.37). I bet most American companies would choose Fable over GPT-5.5. Employee salaries cost far more than token costs. Getting the job done right is much more important. [0]: https://cursor.com/cursorbench

Sure, my point about "things will eventually catch up to, and surpass Fable" still stands.

Re: Claude Sonnet 5

#794

Post by brokencode » Wed, Jul 01, 2026, 5:55 PM UTC

siva7 wrote: post #591

brokencode wrote: post #316

Earlier quoted context omitted.

Kind of crazy how bad this release actually is. I even dug around in the full system card, and every graph showed the same thing. Low and maybe medium will save money on simpler tasks, but after that it just isn’t worth it compared to Opus. I wish they would have explained in the blog post why they think anybody would ever want to use this above medium. Maybe it works well on things that aren’t clear in the benchmark…

Why would a company explain how limited their own major release is?

The graphs do that already. I was expecting them to try to explain how good it was at simple tasks.

Re: Claude Sonnet 5

#795

Post by theshrike79 » Wed, Jul 01, 2026, 6:33 PM UTC

Taek wrote: post #89

kristianc wrote: post #75

Earlier quoted context omitted.

There's two classes of models now - the cybersecurity ones that none of us are getting, and the 'safe' models released for general consumption. This is letting us know which side of the divide it sits on.

There's also Chinese models, which aren't trying to self-limit capabilities.

How do you know, if you use ther API and don't self-host the full model?

Re: Claude Sonnet 5

#796

Post by sanderjd » Wed, Jul 01, 2026, 8:54 PM UTC

whateveracct wrote: post #622

sanderjd wrote: post #191

Earlier quoted context omitted.

Not for a business model that scales revenue by token usage. But other business models are available.

Like?

I dunno. That's the kind of thing the people who come to this site are meant to figure out though :)

(I mean, I have some ideas, but I don't know if they are good.)

Re: Claude Sonnet 5

#797

Post by sanderjd » Wed, Jul 01, 2026, 8:57 PM UTC

drob518 wrote: post #744

sanderjd wrote: post #529

Earlier quoted context omitted.

Yes. But is your impression that this is the strategy of people like Amodei? My impression is that it isn't, that they are actually true believers, and not just trying to hit the timing right and flip it.

I believe that OpenAI is in trouble. Too many high level defectors. Altman is clearly in hype mode. Is Amodei? Yes, I’m pretty sure he is at some level. He might also believe it partially (it’s not totally binary).

I have a different impression of Amodei, but my impression of openai is similar to yours. And fully agreed that it isn't a binary.

Re: Claude Sonnet 5

#798

Post by ricardobeat » Wed, Jul 01, 2026, 9:24 PM UTC

eli wrote: post #422

yieldcrv wrote: post #416

Earlier quoted context omitted.

What’s everyone favorite GLM provider? z.ai doesnt always have the most reliable AI but I don’t mind the party seeing my trade secrets and thoughts compared to an American corporation + the party seeing my trade secrets and thoughts. So thats not a functional difference to me, and the Chinese one won’t reply to subpoenas so thats a value add tbh So I’ll consider all, fastest tokens/sec wins

Fireworks.ai is solid. And if you care more about speed than cost they have a "fast" variant that I think just throws more hardware at the model for about 2x the cost.

Fast variants are usually quantized to NVFP4, which incurs a slight degradation in intelligence.

Re: Claude Sonnet 5

#799

Post by ethbr1 » Thu, Jul 02, 2026, 12:37 AM UTC

post-it wrote: post #289

itopaloglu83 wrote: post #245

Earlier quoted context omitted.

More and more I find myself trying to stop Opus from doing something stupid, and at every turn I need to tell it to stop overcomplicating things. I think the models are being optimized for wealth extraction from users and companies, instead of solving problems. I don't know why Opus would try to create an entire library when I told it specifically to do something simple that would take 2-3 lines of Python.

> I don't know why Opus would try to create an entire library when I told it specifically to do something simple that would take 2-3 lines of Python. Because it reasons in one direction. First it encounters some kind of issue with 2-3 lines of Python that might make it not work, and then it goes onto plan B, which is making a library, but it doesn't circle back and compare the effort of making the library to working…

So, as the quip goes, junior level reasoning.

Re: Claude Sonnet 5

#800

Post by adam_arthur » Thu, Jul 02, 2026, 1:24 AM UTC

grim_io wrote: post #786

adam_arthur wrote: post #760

Earlier quoted context omitted.

This is the same concept as Chain of Thought. Just that when "native" thinking is used it's hidden from the end result via special tags. If you force a model to reason about a result before producing the result, you get more accurate results. Because "reason" comes before the selection, it has to think through why it is producing the result beforehand (e.g. produce a block of text that makes the correct answer statis…

I'm aware how reasoning works. But maybe I am misunderstanding how you use it. Because from what I see, you are not getting a chain of thought, but just a summary of a non-chain-of-thought answer, with no reasoning either explicit or implicit.

If the LLM has to write out the reason, it is "thinking".

Whether you ask it to give or , both will produce similar chain of thought processes.

To explain the "reason" requires producing thoughts that justify the answer that is not produced yet, because it writes the JSON linearly from top to bottom.

Not sure where the misunderstanding is coming from.