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

Claude Opus 4.5

anthropic.com

Post reply on HN
Print view View original

461–470 of 525 posts

Re: Claude Opus 4.5

#461

Post by andai »

sqs wrote: post #186
llamasushi wrote: post #30

The burying of the lede here is insane. $5/$25 per MTok is a 3x price drop from Opus 4. At that price point, Opus stops being "the model you use for important things" and becomes actually viable for production workloads. Also notable: they're claiming SOTA prompt injection resistance. The industry has largely given up on solving this problem through training alone, so if the numbers in the system card hold up under a…

What's super interesting is that Opus is cheaper all-in than Sonnet for many usage patterns. Here are some early rough numbers from our own internal usage on the Amp team (avg cost $ per thread): - Sonnet 4.5: $1.83 - Opus 4.5: $1.30 (earlier checkpoint last week was $1.55) - Gemini 3 Pro: $1.21 Cost per token is not the right way to look at this. A bit more intelligence means mistakes (and wasted tokens) avoided.

Yeah, that's a great point.

ArtificialAnalysis has a "intelligence per token" metric on which all of Anthropic's models are outliers.

For some reason, they need way less output tokens than everyone else's models to pass the benchmarks.

(There are of course many issues with benchmarks, but I thought that was really interesting.)

Re: Claude Opus 4.5

#462

Post by bamboozled »

stavros wrote: post #11

Did anyone else notice Sonnet 4.5 being much dumber recently? I tried it today and it was really struggling with some very simple CSS on a 100-line self-contained HTML page. This never used to happen before, and now I'm wondering if this release has something to do with it. On-topic, I love the fact that Opus is now three times cheaper. I hope it's available in Claude Code with the Pro subscription. EDIT: Apparently…

Noticed it hard today, it's just "stupid" now.

Re: Claude Opus 4.5

#464

Post by ACCount37 »

sillyfluke wrote: post #454
ACCount37 wrote: post #203

Earlier quoted context omitted.

No, it's entirely psychological. Users are not reliable model evaluators. It's a lesson the industry will, I'm afraid, have to learn and relearn over and over again.

I don't really find this a helpful line to traverse. By this line of inquiry most of the things in software are psychological. Whether something is a bug or feature. Whether the right thing was built. Whether the thing is behaving correctly in general. Whether it's better at the very moment that the thing occasionally works for a whole range of stuff or that it works perfectly for a small subset. Whether fast results…

No. I'm saying that if you take the same exact LLM on the same exact set of hardware and serve it to the same exact humans, a sizeable amount of them will still complain about "model nerfs".

Why? Because humans suck.

Re: Claude Opus 4.5

#465

Post by chickensong »

brianjking wrote: post #382
IgorPartola wrote: post #311

Earlier quoted context omitted.

For me this is sub agent usage. If I ask Claude Code to use 1-3 subagents for a task, the 5 hour limit is gone in one or two rounds. Weekly limit shortly after. They just keep producing more and more documentation about each individual intermediate step to talk to each other no matter how I edit the sub agent definitions.

Care sharing some of your sub-agent usage? I've always intended to really make use of them, but with skills, I don't know how I'd separate these in many use cases?

They're useful for context management. I use frequently for research in a codebase, looking for specific behavior, patterns, etc. That type of thing eats a lot of context because a lot of data needs to be ingested and analyzed.

If you delegate that work to a sub-agent, it does all the heavy lifting, then passes the results to the main agent. The sub-agent's context is used for all the work, not the main agent's.

Re: Claude Opus 4.5

#466

Post by NiloCK »

losvedir wrote: post #102
llamasushi wrote: post #30

The burying of the lede here is insane. $5/$25 per MTok is a 3x price drop from Opus 4. At that price point, Opus stops being "the model you use for important things" and becomes actually viable for production workloads. Also notable: they're claiming SOTA prompt injection resistance. The industry has largely given up on solving this problem through training alone, so if the numbers in the system card hold up under a…

I almost scrolled past the "Safety" section, because in the past it always seemed sort of silly sci-fi scaremongering (IMO) or things that I would classify as "sharp tool dangerous in the wrong hands". But I'm glad I stopped, because it actually talked about real, practical issues like the prompt injections that you mention. I wonder if the industry term "safety" is pivoting to refer to other things now.

Waymos, LLMs, brain computer interfaces, dictation and tts, humanoid robots that are worth a damn.

Ye best start believing in silly sci-fi stories. Yer in one.

Re: Claude Opus 4.5

#467

Post by leo_e »

sqs wrote: post #186
llamasushi wrote: post #30

The burying of the lede here is insane. $5/$25 per MTok is a 3x price drop from Opus 4. At that price point, Opus stops being "the model you use for important things" and becomes actually viable for production workloads. Also notable: they're claiming SOTA prompt injection resistance. The industry has largely given up on solving this problem through training alone, so if the numbers in the system card hold up under a…

What's super interesting is that Opus is cheaper all-in than Sonnet for many usage patterns. Here are some early rough numbers from our own internal usage on the Amp team (avg cost $ per thread): - Sonnet 4.5: $1.83 - Opus 4.5: $1.30 (earlier checkpoint last week was $1.55) - Gemini 3 Pro: $1.21 Cost per token is not the right way to look at this. A bit more intelligence means mistakes (and wasted tokens) avoided.

Hard agree. The hidden cost of 'cheap' models is the complexity of the retry logic you have to write around them.

If a cheaper model hallucinates halfway through a multi-step agent workflow, I burn more tokens on verification and error correction loops than if I just used the smart model upfront. 'Cost per successful task' is the only metric that matters in production.

Re: Claude Opus 4.5

#468

Post by UltraSane »

victor9000 wrote: post #131
UltraSane wrote: post #63

Earlier quoted context omitted.

I've had Gemini 3 Pro solve issues that Claude Code failed to solve after 10 tries. It even insulted some code that Sonnet 4.5 generated

I'm also finding Gemini 3 (via Gemini CLI) to be far superior to Claude in both quality and availability. I was hitting Claude limits every single day, at that point it's literally useless.

Hopefully once Anthropic has 1 million Google TPUs in use they will have sufficient capacity.

Re: Claude Opus 4.5

#469

Post by Havoc »

Interesting that the number of hn comments on big model announcements seems to be dropping. I recall previous ones easily surpassing 1k

Maybe models are starting to get good enough/ levelling off?

Re: Claude Opus 4.5

#470

Post by atq2119 »

mannanj wrote: post #329
shepherdjerred wrote: post #113

Earlier quoted context omitted.

I thought AI safety was dumb/unimportant until I saw this dataset of dangerous prompts: https://github.com/mlcommons/ailuminate/blob/main/airr_offic... I don't love the idea of knowledge being restricted... but I also think these tools could result in harm to others in the wrong hands

I once heard a devils advocate say, “if child porn can be fully AI generated and not imply more exploitation of real children, and it’s still banned then it’s about control not harm.” Attack away or downvote my logic.

The counter-devil's advocate[0] is that consuming CSAM, whether real or not, normalizes the behavior and makes it more likely for susceptible people to actually act on those urges in real life. Kind of like how dangerous behaviors like choking seem to be induced by trends in porn.

[0] Considering how CSAM is abused to advocate against civil liberties, I'd say there are devils on both sides of this argument!

Post reply on HN