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

Claude Sonnet 4.5

anthropic.com

Post reply on HN
Print view View original

641–650 of 819 posts

Re: Claude Sonnet 4.5

#641

Post by stravant »

zmmmmm wrote: post #595
Bjorkbat wrote: post #169

> Practically speaking, we’ve observed it maintaining focus for more than 30 hours on complex, multi-step tasks. Really curious about this since people keep bringing it up on Twitter. They mention it pretty much off-handedly in their press release and doesn't show up at all in their system card. It's only through an article on The Verge that we get more context. Apparently they told it to build a Slack clone and left…

> Apparently they told it to build a Slack clone and left it unattended for 30 hours, and it built a Slack clone using 11,000 lines of code it's going to be an issue I think, now that lots of these agents support computer use, we are at the point where you can install an app, tell the agent you want something that works exactly the same and just let it run until it produces it. The software world may find it's got mo…

That just means that owning the walled gardens and network effects will become yet more important.

Re: Claude Sonnet 4.5

#642

Post by artdigital »

blurbleblurble wrote: post #56
yewenjie wrote: post #5

Looking at the chart here, it seems like Sonnet 4 was already better than GPT-5-codex in the SWE verified benchmark. However, my subjective personal experience was GPT-5-codex was far better at complex problems than Claude Code.

GPT-5 is like the guy on the baseball team that's really good at hitting home runs but can't do basic shit in the outfield. It also consistently gets into drama with the other agents e.g. the other day when I told it we were switching to claude code for executing changes, after badmouthing claude's entirely reasonable and measured analysis it went ahead and decided to `git reset --hard` even after I twice pushed back…

Gemini is an excellent collaborator?

It’s the one AI that keeps telling me I’m wrong and refuses to do what I ask it to do, then tells me “as we have already established, doing X is pointless. Let’s stop wasting time and continue with the other tasks”

It’s by far the most toxic and gaslighting LLM

Re: Claude Sonnet 4.5

#643

Post by dwaltrip »

martin-t wrote: post #583
int_19h wrote: post #541

Earlier quoted context omitted.

It's not that it was uneconomical, it's that 1) we literally don't know all the rules, a lot of it is learned intuition that humans acquire by doing , and 2) as task complexity rises, the number of rules rises faster, so it doesn't scale. The real advantage that genAI brings to the table is that it "learns" in a way that can replicate this intuition and that it keeps scaling so long as you can shovel more compute and…

In a way, yes, you'd be paying the people not just to write down the rules but to discover them first. And there's the accuracy/correctness/interpretability tradeoff. But also, have there been any attempts on the scale of the Manhattan project attempting to create a GOFAI? Because one idea I ran into is that we might be able to use genAI to create a GOFAI soon. And it would be as hard as using genAI for any kind of l…

You can’t “write down the rules” for intelligence. Not for any reasonable definition of “writing”. The medium of writing is not rich enough to express what is needed.

This is why GOFAI failed.

Re: Claude Sonnet 4.5

#644

Post by mirsadm »

lexarflash8g wrote: post #293

Just tested this on a rather simple issue. Basically it falls into rabbits holes just like the other models and tries to brute force fixes through overengineering through trial and error. It also says "your job should now pass" maybe after 10 prompts of roughly doing the same thing stuck in a thought loop. A GH actions pipeline was failing due to a CI job not having any source code files -- error was "No build system…

They're very good at things have been done a million times before. I use both Claude and Gemini and they are pretty terrible at writing any kind of Vulkan shader but really good for spitting out web pages and small bits of code here and there. For me that's enough to make them useful.

Re: Claude Sonnet 4.5

#645

Post by mirsadm »

labrador wrote: post #557
AbuAssar wrote: post #432

I used to treat writing code as a form of art, with attention to details and best practices, and using design patterns whenever possible. but it seems this will come to an end eventually as these agents become more stronger and capable each day, and will be better and faster than human coders.

I'm sympathetic, but it occcured to me that ccording to my amatuer studies, Germany lost WW2 in part because it had a craftsman mentality to manufacture war machines and ended up with a bazillion different part requirments and a shortage of skilled craftsmen, while America used Henry Ford's assembly line process to stamp out hundreds of thousands of identical machines sharing the same parts. Now we are at the assembl…

This has been said about the release of every model in the last couple of years. Personally I can't even tell if this one is better than 3.5. I actually found that one more useful.

Re: Claude Sonnet 4.5

#646

Post by coolfox »

manofmanysmiles wrote: post #521

I haven't shouted into the void for a while. Today is as good a day as any other to do so. I feel extremely disempowered that these coding sessions are effectively black box, and non-reproducible. It feels like I am coding with nothing but hopes and dreams, and the connection between my will and the patterns of energy is so tenuous I almost don't feel like touching a computer again. A lack of determinism comes from m…

> A lack of determinism comes from many places, but primarily: 1) The models change 2) The models are not deterministic...

models themselves are deterministic, this is a huge pet peeve of mine, so excuse the tangent, but the appearance of nondeterminism comes from a few sources, but imho can be largely attributed to the probabilistic methods used to get appropriate context and enable timely responses. here's an example of what I mean, a 52-card deck. The deck order is fixed once you shuffle it. Drawing "at random" is a probabilistic procedure on top of that fixed state. We do not call the deck probabilistic. We call the draw probabilistic. Another exmaple, a pot of water heating on a stove. Its temperature follows deterministic physics. A cheap thermometer adds noisy, random error to each reading. We do not call the water probabilistic. We call the measurement probabilistic.

Theoretical physicists run into such problems, albeit far more complicated, and the concept for how they deal with them is called ergodicity. The models at the root of LLM's do exhibit ergodic behavior; the time average and the ensemble average of an observable are identical, i.e. the average response of a single model over a long duration and the average of many similar models at a fixed moment are equivalent.

Re: Claude Sonnet 4.5

#647

Post by 827a »

manofmanysmiles wrote: post #521

I haven't shouted into the void for a while. Today is as good a day as any other to do so. I feel extremely disempowered that these coding sessions are effectively black box, and non-reproducible. It feels like I am coding with nothing but hopes and dreams, and the connection between my will and the patterns of energy is so tenuous I almost don't feel like touching a computer again. A lack of determinism comes from m…

Yeah, tbh I used to be a bit agentic coding tool-pilled, but over the past four months I've come to realize that if this industry evolves in a direction where I don't actually get to write code anymore, I'm just going to quit.

Code is the only good thing about the tech industry. Everything else is capitalist hellscape shareholder dystopia. Thinking on it, its hilarious that any self-respecting coder is excited about these tools, because what you're excited for is a world where, now, at best, your entire job is managing unpredictable AI agents while sitting in meetings all day to figure out what to tell your AI agents to build. You don't get to build the product you want. You don't get to build it how you want. You'll be a middle manager that gets to orchestrate the arguments between the middle manager you already had and the inflexible computer.

You don't have to participate in a future you aren't interested in. The other day my boss asked me if I could throw Cursor at some task we've had backlogged for a while. I said "for sure my dude" then I just did it myself. It took me like four hours, and my boss was very impressed with how fast Cursor was able to do it, and how high quality the code was. He loves the Cursor metrics dashboard for "lines accepted" or whatever, every time he screenshares he has that tab open, so sometimes I task it on complicated nonsense tasks then just throw away the results. Seeing the numbers go up makes him happy, which makes my life easier, so its a win-win. Our CTO is really proud of "what percentage of our code is AI written" but I'm fairly certain that even the engineers who use it in earnest actually commit, like, 5% of what Cursor generates (and many do not use it in earnest).

The sentiment shift I've observed among friends and coworkers has been insane over the past two months. Literally no one cares about it anymore. The usage is still there, but its a lot more either my situation or just a "spray and pray" situation that creates a ton of disillusioned water cooler conversations.

Re: Claude Sonnet 4.5

#648

Post by baq »

stefs wrote: post #272
itsoktocry wrote: post #197

Earlier quoted context omitted.

> It’s a simple substitution request where I provide a Lint error that suggests the correct change. All the models fail. I could ask someone with no development experience to do this change and they could. I don't understand why this kind of thing is useful. Do the thing yourself and move on. For every one problem like this, AI can do 10 better/faster than I can.

How can I trust it to do the complicated task well when it fails to do the simple thing?

The jagged edge effect: you can trust it to do some tasks extremely well, but a slightly different task might consistently fail. Your job as a tool user is to understand when it’ll work and when it won’t - it isn’t an oracle or a human.

Re: Claude Sonnet 4.5

#649

Post by baq »

Marazan wrote: post #133
blurbleblurble wrote: post #110

Earlier quoted context omitted.

Yeah, it's wild how the biases get encoded in there. Maybe they aren't even entirely separable from the magic of LLMs.

It isn't wild, it is inherent to the very nature of large language models. The power of using LLMs is working out what it has encoded and how to access it.

It’s as if we made the machine in our own image. Who would’ve thought /s

Perhaps for the first time in history we have to understand culture when working with a tool, but it’s still just a tool.

Re: Claude Sonnet 4.5

#650

Post by baq »

jasonsb wrote: post #37
yewenjie wrote: post #5

Looking at the chart here, it seems like Sonnet 4 was already better than GPT-5-codex in the SWE verified benchmark. However, my subjective personal experience was GPT-5-codex was far better at complex problems than Claude Code.

My subjective personal experience is the exact opposite of yours, GPT-5-codex is super slow and the results are mediocre at best. I would probably stop using AI for coding if I was forced to use GPT-5-codex.

GPT-5-high (haven’t tried codex yet) is dog slow, but IME if you start with asking it for detailed requirements in a markdown doc with alternatives for each major decision and pseudocode implementations with references to relevant files, it makes a great prompt for faster a model like sonnet.
Post reply on HN