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

411–420 of 819 posts

Re: Claude Sonnet 4.5

#411

Post by atonse »

inopinatus wrote: post #391
siva7 wrote: post #380

Does 4.5 still answer everything with "You're absolutely right!" or is it now able to communicate like a real programmer?

I won’t be satisfied until I get a Linus Torvalds mode. “Your idea is shit because you are so fucking stupid” “Please stop talking, it hurts my GPUs thinking down to your level” “I may seem evil but at least I’m not incompetent”

Why is this getting downvoted? It was hilarious!

I actually added a fun thing to my user-wide CLAUDE.md, basically saying that it should come up with a funny insult every time I come up with an idea that wasn't technically sound (I got the prompt from someone else). It seems to be disobeying me, because I refuse to believe that I don't have bad ideas. Or some other prompt is overriding it.

Re: Claude Sonnet 4.5

#413

Post by Szpadel »

iagooar wrote: post #395
dbbk wrote: post #324

Earlier quoted context omitted.

Did you use plan mode?

Yes, I did. I ran the test again, took Claude ~4mins this time. There was no error now with the auth, but the functionality was totally broken. It could not even find the most basic stuff that matches perfectly.

try also adding Serena mcp and make sure you did /init

Re: Claude Sonnet 4.5

#414

Post by epolanski »

iagooar wrote: post #258

Anecdotal evidence. I have a fairly large web application with ~200k LoC. Gave the same prompt to Sonnet 4.5 (Claude Code) and GPT-5-Codex (Codex CLI). "implement a fuzzy search for conversations and reports either when selecting "Go to Conversation" or "Go to Report" and typing the title or when the user types in the title in the main input field, and none of the standard elements match, a search starts with a 2s de…

There's few issues with posts like yours:

1. Different LLMs require different prompts and information

2. They ignore LLMs non determinism, you should run the experiment several times

Re: Claude Sonnet 4.5

#415

Post by rossjudson »

Implicated wrote: post #274
iagooar wrote: post #258

Anecdotal evidence. I have a fairly large web application with ~200k LoC. Gave the same prompt to Sonnet 4.5 (Claude Code) and GPT-5-Codex (Codex CLI). "implement a fuzzy search for conversations and reports either when selecting "Go to Conversation" or "Go to Report" and typing the title or when the user types in the title in the main input field, and none of the standard elements match, a search starts with a 2s de…

I'm not trying to be offensive here, feel the need to indicate that. But that prompt leads me to believe that you're going to get rather 'random' results due to leaving SO much room for interpretation. Also, in my experience, punctuation is important - particularly for pacing and grouping of logical 'parts' of a task and your prompt reads like a run on sentence. Making a lot of assumptions here - but I bet if I were…

When asking for change, there are the choices you know about and the ones you don't. I've gotten in the habit of describing some of the current state as well as my desired state, and using that to focus the LLM on the areas I'd like to have a stronger voice in.

Of course, I definitely appreciate when it makes choices that I don't know I need to make, and it chooses reasonable defaults.

I mean, I could say "make the visualization three columns", but there's a lot of ambiguity in that kind of thing, and the LLM is going to make a lot of choices about my intent.

Instead, "each team/quarter currently has a vertically stacked list of people assigned to that team, with two columns (staying on team, leaving team). change that to being three columns (entering team, staying on team, leaving team)."

As a bonus, it's much, much clearer to somebody reading the history later what the intent was.

Re: Claude Sonnet 4.5

#416

Post by epolanski »

Aflynn50 wrote: post #374

When I see how much the latest models are capable of it makes me feel depressed. As well as potentially ruining my career in the next few years, its turning all the minutiae and specifics of writing clean code, that I've worked hard to learn over the past years, into irrelivent details. All the specifics I thought were so important are just implementation details of the prompt. Maybe I've got a fairly backwards view…

That minutiae was always borderline irrelevant, the skill was always making somebody money, possibly with software.

The reality is that more software will be pushed than before, and more of it will need to be overseen by a professional.

Re: Claude Sonnet 4.5

#417

Post by blurbleblurble »

Sammi wrote: post #387
doctoboggan wrote: post #95

Earlier quoted context omitted.

I do it as well. I have a Claude code instance running in my backend repo, and one running in my frontend repo. If there is required coordination, I have the backend agent write a report for the front end agent about the new backend capabilities, or have the front end agent write a report requesting a new endpoint that would simplify the code. Lots of other people also follow the architect and builder pattern, where…

Sure. But at no point do you need to talk about the existence of other agents. You talk about making a plan, and you talk about implementing the plan. There's no need to talk about where the plan came from.

Because the plan involves using multiple agents with different roles and I don't want them conflicting.

Sure there's no need to explicitly mention the agents themselves, but it also shouldn't trigger a pseudo-jealous panic with trash talk and a sudden `git reset --hard` either.

And also ideally the agents would be aware of one another's strengths and weaknesses and actually play to them rather than sabotaging the whole effort.

Re: Claude Sonnet 4.5

#418

Post by rtp4me »

Just updated to Sonnet 4.5 and Claude Code 2.0 this afternoon. I worked on a quick project (creating PXE bootable files) using the updates and have to say, this new version seems much faster and more accurate than before. I did not go round-and-round trying to get good output and Claude did not go down rabbit holes like before. So far, so good.

Re: Claude Sonnet 4.5

#419

Post by landl0rd »

nchmy wrote: post #341
asadotzler wrote: post #315

Earlier quoted context omitted.

AI blogger seems more appropriate than journalist.

are you aware of any "ai journalists"? Because simonw does great work, so perhaps blogger is what people should aspire towards?

Kylie Robison recently moved to Wired and is a solid "AI journalist".

Re: Claude Sonnet 4.5

#420

Post by epolanski »

CuriouslyC wrote: post #9
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.

The Anthropic models have been vibe-coding tuned. They're beasts at simple python/ts programs, but they definitely fall apart with scientific/difficult code and large codebases. I don't expect that to change with the new Sonnet.

They are very good with C too, but it helps that there's gazzilions of lines of C out there.
Post reply on HN