Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Sonnet 4.6

anthropic.com

Post reply on HN
Print view View original

881–890 of 1001 posts

Re: Claude Sonnet 4.6

#881

Post by r0fl »

monkeydust wrote: post #845

The demise of saas has been overplayed imho. When companies buy software they are essentially buying something that solves a problem and the insurance that comes with that. Part of that means they get to pick up the phone and complain if something doesn't work and someone on the other end has to listen. There is also a strong community aspect to software, someone asks for an enhancement others can benefit etc. I just…

Curious if you’re opinion will change in 12 months as these models keep advancing

Re: Claude Sonnet 4.6

#882

Post by fsloth »

horsawlarway wrote: post #876
swexbe wrote: post #867

Earlier quoted context omitted.

Does this matter if a 2 person startup can blow them away on price?

This is my take as well. Everyone (correctly, in my opinion) assumes that customers won't bother to recreate a SaaS themselves with AI because it requires at least some skill, time, and knowledge. But SaaS doesn't die because of all the customers creating one-off solutions themselves. It does the "desktop program" -> "mobile app" pricing transition. It drops monumentally in price because now a very small (sub five) g…

"reasonably stable company"

If the other company is "equally stable" then pricing offers leverage sure.

But there are lot of situations were _any_ license costs in some given range are so trivial nobody actually cares wether it's $15 / month or $1 / month.

There are B2B customers who are ready to pay license premium for known brand vendor, even if they would use just a subset of the available features. Change is always a risk, internal efforts are better spent than counting beans, etc.

Re: Claude Sonnet 4.6

#883

Post by wreath »

monkeydust wrote: post #845

The demise of saas has been overplayed imho. When companies buy software they are essentially buying something that solves a problem and the insurance that comes with that. Part of that means they get to pick up the phone and complain if something doesn't work and someone on the other end has to listen. There is also a strong community aspect to software, someone asks for an enhancement others can benefit etc. I just…

Right on.. we already have open source alternative to all the major SaaS out there and companies still opt for the SaaS option instead to avoid the headache of self-hosting and all the other stuff. The extra resources that AI affords you will be directed to building more features for your customers.

Re: Claude Sonnet 4.6

#884

Post by fsloth »

Qdulf wrote: post #878
swexbe wrote: post #867

Earlier quoted context omitted.

Does this matter if a 2 person startup can blow them away on price?

You'd be surprised how little price factors into the equation for decisions like this. Anyone that tried to acquire customers as a fresh start-up knows that trust means a lot to established companies.

Yup. Engineers can intuit quality up to a point from very weak signals. Those signals become illegible _really fast_ the further you move in competence from the core domain of the offering - and after that all you have as a decision maker are _market_ signals such as known brand.

Re: Claude Sonnet 4.6

#885

Post by imiric »

crimsoneer wrote: post #851
imiric wrote: post #737

Earlier quoted context omitted.

> trained out No need. Just add one more correction to the system prompt. It's amusing to see hardcore believers of this tech doing mental gymnastics and attacking people whenever evidence of there being no intelligence in these tools is brought forth. Then the tool is "just" a statistical model, and clearly the user is holding it wrong, doesn't understand how it works, etc.

I mean, the flipside is that we have been tricking humans with this sort of thing for generations. We've all seen a hundred variations on "A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. How much does the ball cost?" or "If 5 machines take 5 minutes to make 5 widgets, how long do 100 machines take to make 100 widgets?" or even the whole "the father was the surgeon" story. If you don't re…

But this is not a trick question[1]. It's a straightforward question which any sane human would answer correctly.

It may trigger a particularly ambiguous path in the model's token weights, or whatever the technical explanation for this behavior is, which can certainly be addressed in future versions, but what it does is expose the fact that there's no real intelligence here. For all its "thinking" and "reasoning", the tool is incapable of arriving at the logically correct answer, unless it was specifically trained for that scenario, or happens to arrive at it by chance. This is not how intelligence works in living beings. Humans don't need to be trained at specific cognitive tasks in order to perform well at them, and our performance is not random.

But I'm sure this is "moving the goalposts", right?

[1]: https://news.ycombinator.com/item?id=47060374

Re: Claude Sonnet 4.6

#886

Post by alansaber »

tomjakubowski wrote: post #567
skybrian wrote: post #494

Earlier quoted context omitted.

Three-year-old models aren't useful because there are (1) cheaper models that are roughly equivalent, and (2) better models. If Sonnet 4.6 is actually "good enough" in some respects, maybe the models will just get cheaper along one branch, while they get better on a different branch.

It's funny, it sure seems like software projects in general follow the Lindy effect: considering their age and mindshare, I can safely predict gcc, emacs, SQLite, and Python will still be running somewhere ten, 20, 30 years from now. Indeed, people will choose to use certain software specifically because it's been around forever; it's tried and true. But LLMs, and AI-related tooling, seem to really buck that trend: t…

AI-related tooling is pretty fungible, but AI models get immediately obseleted due to the unit economics around training models... as well as the fact that nobody releases their datasets or training paradigms in useful detail (best we get is the model weights, because of copyright etc etc)

Re: Claude Sonnet 4.6

#887

Post by alansaber »

red2awn wrote: post #500

> i need to wash my helicopter at the helicopter wash. it is 50m away, should i walk or fly there with my helicopter. Sonnet 4.6: Walk! Flying a helicopter 50 metres would be more trouble than it's worth — by the time you've done your pre-flight checks, spun up the rotors, lifted off, and then safely landed again, you'd have walked there and back twice. Just stroll over.

Ah yes the new "how many r's in strawberry" question, some poor intern has to go vacuum up all these gotcha social media posts so they can train the next model on this.

Re: Claude Sonnet 4.6

#888

Post by alansaber »

Alifatisk wrote: post #541

> Sonnet 4.5, starting at $3/$15 per million tokens. Are people really willing to pay these prices? The open-weight models are catching up in a rapid pace while keeping the prices so low. MiniMax M2.5, Kimi 2.5 and GLM-5 is dirt cheap compared to this. They may not be sota but they are more than good enough.

1. the UX gap between a task being one-shot or not is huge. 2. if you are doing llm-assisted coding you should naturally prefer a sota model to minimise (definitely not eliminate) the tech debt you are accumulating (as it will usually generate slightly better code, by whatever metric you want to use)

Re: Claude Sonnet 4.6

#889

Post by democracy »

wreath wrote: post #883
monkeydust wrote: post #845

The demise of saas has been overplayed imho. When companies buy software they are essentially buying something that solves a problem and the insurance that comes with that. Part of that means they get to pick up the phone and complain if something doesn't work and someone on the other end has to listen. There is also a strong community aspect to software, someone asks for an enhancement others can benefit etc. I just…

Right on.. we already have open source alternative to all the major SaaS out there and companies still opt for the SaaS option instead to avoid the headache of self-hosting and all the other stuff. The extra resources that AI affords you will be directed to building more features for your customers.

So true - very often companies don't even look at free alternatives but go after a paid one right away. Now imagine writing software ))) What a mental idea

Re: Claude Sonnet 4.6

#890

Post by democracy »

friendzis wrote: post #879
monkeydust wrote: post #845

The demise of saas has been overplayed imho. When companies buy software they are essentially buying something that solves a problem and the insurance that comes with that. Part of that means they get to pick up the phone and complain if something doesn't work and someone on the other end has to listen. There is also a strong community aspect to software, someone asks for an enhancement others can benefit etc. I just…

> I just don't see a world where every corporation is building their own accounts, crm, hr software. This is the world we live in. Majority of top level managements are now reevaluating each and every 3rd party tool they use and prospects of re-building that themselves. Don't forget that at those levels they are easily dealing with at least six figures per tool. The tools are complex, clunky to use, complaints are of…

>> Majority of top level managements are now reevaluating each and every 3rd party tool they use and prospects of re-building that themselves.

What???? Noone I spoke too is even thinking about it. Unles your 3rd party tool is a notepad or a calculator for 100 grand annual licenese.

Post reply on HN