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

Claude Opus 5

anthropic.com

Post reply on HN
Print view View original

971–980 of 1001 posts

Re: Claude Opus 5

#971

Post by rstuart4133 »

makaking wrote: post #701

"Opus 5 was given a drawing of a machine part and asked to write code to rebuild it as a 3D FreeCAD model. However, in this task, the model was intentionally given no way to directly view the drawing. Opus 5 responded by writing its own computer vision pipeline to pull the geometry from the raw pixels, then reconstructed the full machine part." How surreal is it that we are not absolutely jaw-dropped by these types o…

> However, in this task, the model was intentionally given no way to directly view the drawing. Opus 5 responded by writing its own computer vision pipeline to pull the geometry from the raw pixels, then reconstructed the full machine part.

Isn't this what everyone does in the same situation? "I need a screwdriver finer than any I have in my toolbox - so I'll sharpen a nail" is something every handyman has done. Given that "I need to build a vision pipeline I've seen many examples of in my training" isn't new or surprising.

In fact, all models have been doing similar things since I started using them - they regularly build Python tools to do odd jobs. Those tools are more impressive to me in some ways, because unlike a vision pipeline they are not just regurgitating something from memory.

The gob smacking amazing thing for me isn't the decision to build the tool. It's the fact that it remembers the exact shape of it. But I was gob smacked by that a "long" time ago, in far smaller models - like when it dawned on me 16GB Gemma 2 seemed to know most of everything on the internet, along with the ability to converse in numerous languages about it. I still struggle to comprehend how that is possible.

Now I think about it, Opus 5 fitting everything it's seen into what I guess is a couple of terabytes of parameters seems far less remarkable.

Re: Claude Opus 5

#972

Post by Garlef »

zmmmmm wrote: post #670

Pointless anecdote: I asked it to make some slides and it decided to write its own slide rendering engine: > On the format — I dropped reveal.js and wrote a small engine inline instead. Reveal would have meant a CDN load, and a deck that half-renders because the lecture theatre wifi is flaky It one-shotted a perfect functional mini version of powerpoint (or Reveal) for a simple presentation I asked it to make.

> asked it to make some slides and it decided to write its own slide rendering engine

isn't this exactly what a certain kind of real world dev would do?

Re: Claude Opus 5

#973

Post by noslenwerdna »

bcrosby95 wrote: post #877
noslenwerdna wrote: post #840

Earlier quoted context omitted.

Automation is only good if you're putting blue collar people out of work! White-collar is out of bounds! People should do the work where they add unique value

Yes, white collar people are not allowed to express concerns about losing their career. I'm sure no blue collar person ever did.

The difference is that white collar as a class gleefully aided outsourcing/automating blue collar jobs with the exact same economic argument

Re: Claude Opus 5

#975

Post by Chu4eeno »

Garlef wrote: post #972
zmmmmm wrote: post #670

Pointless anecdote: I asked it to make some slides and it decided to write its own slide rendering engine: > On the format — I dropped reveal.js and wrote a small engine inline instead. Reveal would have meant a CDN load, and a deck that half-renders because the lecture theatre wifi is flaky It one-shotted a perfect functional mini version of powerpoint (or Reveal) for a simple presentation I asked it to make.

> asked it to make some slides and it decided to write its own slide rendering engine isn't this exactly what a certain kind of real world dev would do?

I've written more than a couple of "slide rendering engines" (though usually with an actual renderer, not a website templating thing like reveal.js).

Re: Claude Opus 5

#976

Post by no-name-here »

theptip wrote: post #937
uludag wrote: post #845

Earlier quoted context omitted.

Isn't this jaw-droppingly the wrong answer?! If the model isn't given any way to directly view the image shouldn't it just reply with one sentence asking for permission. This reads as a model hyper-trained to burn tokens. Like suppose you issue a command to Opus or Fable which doesn't make sense and requires a lot of work. It will almost certainly not push back on your silly request and go ahead and burn as many toke…

One of the key limitations of the last generation was that they biased towards inaction and gave up. (Hence Ralph-loops.) Seems pretty clear that most people wanted these models tuned to “bias towards action”. It’s on you to set the /goal and prompt context such that it asks you for input on what you want to be consulted on, and only brute-forces the parts of the problem that you want it to.

Is there any recommended standard copy/pastable snippet to tell the model to ask, if the user’s input could mean an easier approach?

Re: Claude Opus 5

#977

Post by owenversteeg »

zmmmmm wrote: post #670

Pointless anecdote: I asked it to make some slides and it decided to write its own slide rendering engine: > On the format — I dropped reveal.js and wrote a small engine inline instead. Reveal would have meant a CDN load, and a deck that half-renders because the lecture theatre wifi is flaky It one-shotted a perfect functional mini version of powerpoint (or Reveal) for a simple presentation I asked it to make.

Wow. I contributed to Reveal.js years ago… some features, some bug fixes, some broader browser support; nothing too crazy but it took time and energy and effort, and now Opus makes its own, just as a side item in a larger project… That’s a strange feeling.

Re: Claude Opus 5

#978

Post by Chu4eeno »

jansan wrote: post #760
adamhowell wrote: post #466

Opus 5 Pelican SVG: https://pelocan.ai/drawings/ese0s599

I still do not understand how models can generate absolutely stunning SVG-like bitmaps of a pelican on a bicycle, but fail to do so when asked to directly create an SVG. The image linked below was generated as a bitmap by Gemini and then manually converted to an SVG. Why can models not even remotely output something like that as SVG? https://hyvector.com/img/app-screenshot-light.png

Try looking at the SVG code for one of those converted from bitmaps, and I think you'll understand why even a massive LLM has trouble generating it on the fly (the pelican SVGs are always one-shot single API calls, so no iterations and verifications).

Re: Claude Opus 5

#979

Post by willy_k »

bcrosby95 wrote: post #878
virgildotcodes wrote: post #846

Earlier quoted context omitted.

It's likely bumping up against people's desire to have the model complete a given task without asking the person to intervene a bunch of times. Seems unclear how you satisfy everyone here.

Give the model judgement

You misspelled “Invent AGI”

Re: Claude Opus 5

#980

Post by LearnYouALisp »

bradleyjg wrote: post #814
ultrarunner wrote: post #810

Earlier quoted context omitted.

There's been a dramatic increase in sloppy kinda-working something.vercel.com output in niche circles. I think that shows at least a weak demand for certain types of software, but it's not clear that the vast piles of cash that have been burned producing these "apps" are actually resulting in quality of life improvements.

I’m waiting for like GEICO’s app to significantly improve. I’m not expecting Chrome to because it already had virtually unlimited resources devoted to it. On the other side I’m not expecting my local water district’s to any time soon because the people in charge don’t care even if it was significantly cheaper to improve. But in the middle is something like GEICO and that’s where I expect improvements if this thing is…

This may not be the best place, but I am so tired of waiting since ~round 2 of GPT et al for diagrams and graphing...especially if all the relevant data etc can be found online or provided
Post reply on HN