Top
Best
New

Posted by alvis 4 hours ago

Claude Opus 5(www.anthropic.com)
https://www.anthropic.com/claude-opus-5-system-card
870 points | 483 commentspage 2
acmnrs 4 hours ago|
From the prompting guide<https://platform.claude.com/docs/en/build-with-claude/prompt...>:

> Claude Opus 5's default user-facing responses run longer than prior Opus models'.

The benchmarks do show Opus 5 as slightly more expensive than 4.8, although the scores are much higher.

This still feels like a step in the wrong direction, though, especially with OpenAI making so much progress with the efficiency of their models. Fable's token efficiency made it seem like Anthropic would start following OpenAI's approach but that doesn't seem to have carried over to their other models.

lukeschlather 3 hours ago||
On a small, easily digestible task, I compared Fable to Opus and the cost of Fable was easily 2x despite being fewer tokens, and the output was not really better. Obviously, there are tasks where using Fable matters but honestly they're rather unusual. And for a lot of tasks I've found downgrading to Sonnet can be valuable because Fable and Opus are a lot more secretive about what they're doing, and it's impossible to "listen to them think" and stop them when they start making off-the-wall inferences/assumptions and going down bad paths.

I think in the long run tokens are probably the wrong thing; it's compute and cache memory that you need to be measuring, and when you look at it that way I suspect in most cases the models have pretty similar performance.

adamtaylor_13 3 hours ago|||
This is part of why I switched to Grok 4.5

I don't need more powerful models, I need one that responds fast enough that my attention doesn't wander to other tasks. Grok 4.5 is so fast I can just use it in-band without swapping to other tasks.

Slower than Opus 4.8, which was already miserably slow, is indeed a step in the wrong direction.

MarceColl 1 hour ago||
for you, I am in general not constrained by the speed of models since I parallelize. For me autonomy and accuracy are paramount above all else.
overfeed 3 hours ago|||
> This still feels like a step in the wrong direction, though, especially with OpenAI making so much progress with the efficiency of their models

Gemini also had modest increase before this - don't be surprised when OpenAI also has a "modest increase" with its next release. Cartel-like behaviour doesn't require direct communication when none of the participants are interested in participating in a margin-destroying price-war. All one needs to do is raise their price and watch how the competition react.

Such a scheme (and resulting high margins) would be imperilled by the existence of frontier open-weight models in the market, which may be why the reaction to Chinese models may be particularly shrill.

simianwords 3 hours ago||
>Don't be surprised when OpenAI also has a "modest increase" with its next releases; cartel-like behaviour doesn't require overt coordination when none of the participants are interested in participating in a margin-destroying price-war.

No I will be surprised and I'll bet on the fact that prices will keep going down, just like it went ~50% down in the latest GPT 5.6 release.

holtkam2 3 hours ago|||
The final user-facing responses are usually a tiny fraction of the total tokens used over the course of a given conversation turn. When you're doing any real work, reasoning and tool uses constitute the overwhelming majority of the tokens in / out... not the final user-facing response.
alansaber 4 hours ago|||
Is that true? Sol responses are also longer than prior models.
edumucelli 4 hours ago||
Anecdata: my workflow has been working on the same personal projects for months now with Codex. I cannot anymore finish my daily/weekly code with 4.8 anymore.

I was dividing my work between Codex and DeepSeek. Now I barely use DeepSeek, or never because Codex quota is enough after Sol

elbear 3 hours ago|||
I've had the same experience but at the same time I also find Sol's answers in conversation longer.
wahnfrieden 3 hours ago||||
I hit Codex limits (20x account, never using /fast) on Sol Medium in about 2.5 days
copperx 3 hours ago|||
Which plan?
jjcm 3 hours ago|||
> especially with OpenAI making so much progress with the efficiency of their models

To be fair though, Sol tends to go off the rails sometimes. It's much less reliable than Fable in its outputs. It tends to be overzealous in its research/changes.

aesthesia 2 hours ago|||
This is a comment about user-facing responses, which are seldom the thing you're worried about when thinking about token efficiency.
winwang 1 hour ago|||
Almost completely disagree. Slightly more expensive, but significant better on a per-prompt basis? For non-trivial projects, the former is a small linear increase, the latter is a (somewhat-)exponential(-ish) cost/time/sanity savings.
frenchie4111 4 hours ago|||
It's a step in the wrong direction but also token efficiency has become a focus relatively recently (just the past few weeks it seems like the zeitgeist has turned it's attention to efficiency) while work on this model probably started many many months ago. I would expect to see models released that focus on token efficiency in 6-12 months
theHocineSaad 2 hours ago||
Opus 5 is considered the most intelligent model[0], while it's half the price of Fable 5[1], and Anthropic is still positioning Fable 5 as the most capable model[2].

Is it because maybe Anthropic engineered Opus 5 to work well on benchmarks and didn't do the same thing to Fable 5, or is there another reason?

[0]: https://artificialanalysis.ai/#intelligence

[1]: https://platform.claude.com/docs/en/about-claude/pricing

[2]: https://platform.claude.com/docs/en/about-claude/models/over...

pietz 1 hour ago||
Benchmarks have gotten great, but they're still a proxy for the real world. The 3 GPT 5.6 models are also further apart in reality than the numbers suggest. That said, I'm still mighty impressed how good Luna is for the price. Highly underrated model.

I have been trying to build something that captures the behavioral element of different models, but it's kinda tough.

dgellow 2 hours ago||
That’s what I understand looking at what has been released, but it’s not really clear. The pricing is lower than I expected, I’m wondering what their margin is
not_a9 4 hours ago||
> Opus 5’s safeguards match those of Claude Fable 5’s, with one change: it now permits source-code vulnerability discovery at all access levels. This means that the model can support defensive cybersecurity work while still blocking vulnerability discovery in compiled binaries, which is more commonly used offensively.

Okay so it’s worse than Opus 4.8 for my purposes I guess?

sebzim4500 4 hours ago||
Presumably it drops back to 4.8 in those cases so it's not really worse
tyre 3 hours ago||
yes. At the bottom of the release post it says that they are releasing two new features, one of which is customizing fallback behavior instead of blocking for restricted models
yukIttEft 3 hours ago||
What are your purposes?
not_a9 3 hours ago||
Reversing for the most part, though lately I’ve been doing some code obfuscation/binary rewriting stuff. Fable will switch to Opus instantly on these and I’m unsure how this will perform. I suppose the only way to find out is to test.
roboyoshi 1 hour ago|||
It's rather broad right now. I started reverse engineering a mac app and it started reading some binary data and then quickly told me to switch to Opus 4.8 to continue, because the guardrails kicked in.
ciefa 1 hour ago||||
Try GPT 5.6 Sol if you haven't yet.

I recently created a patch for Riftborne via static IL patching and Fable 5 outright kept refusing to do it, no issue whatsoever with GPT 5.6 Sol lol.

jaggederest 2 hours ago|||
I am very excited for a future where all software is by default modifiable, even shipped binaries, via patches or trampolining, or trickery I don't even know the name of.
derac 43 minutes ago||
It's essentially here, I've had success with gpt-5.6 and Ghidra MCP
atraac 4 hours ago||
Great that there's a new model but they could fix their existing infra. We're considering dropping our Claude Team sub cause it's unusable recently. Constant bugs, dropped sessions, issues switching models, http errors. It's becoming ridiculous
hoppp 4 hours ago||
Funny that a company selling an AI software developer can't use it to fix their infra.

Fixing those issues still requires humans.

websap 4 hours ago|||
How many companies at the size of Anthropic can serve the amount of traffic and manage the amount of compute they have?
amelius 3 hours ago|||
It doesn't matter. Front end code should not misbehave if servers can't keep up. At worst it should fail gracefully.
JackSlateur 3 hours ago||||
How many compagnies can manage a mostly stateless workload at "whatever-the-scale-because-it-does-not-matter-because-stateless" ? Lots of people can do that. Massive amount of people can do that.
artursapek 3 hours ago|||
HN users are world champions are trivializing difficult things with snarky comments
mypalmike 3 hours ago|||
I mean, it's Anthropic‘s front end, Michael. What could it cost? 10 dollars?
pton_xd 3 hours ago|||
I mean, supposedly software engineering is solved so it's somewhat justified snark.
SketchySeaBeast 4 hours ago||||
Let's be honest - they're also still hiring software devs. AI still requires skilled humans in the loop and that's not going away.
arm32 4 hours ago|||
But the job itself may not exist in a year, according to the job posting page.
SketchySeaBeast 3 hours ago|||
There should be a sign somewhere:

934 days since people first started threatening that devs would be replaced by AI in 365 days. 0 day(s) since Anthropic posted a developer job posting.

Only one of those numbers would need to be dynamic.

subscribed 2 hours ago|||
In my long and quite happy and successful career in the adjacent field none of the actually good jobs were the posted.

Specialist headhunters handle that.

moonu 3 hours ago|||
I'm pretty sure that was a fake screenshot made as a joke, not an actual job posting
arm32 3 hours ago||
Nah, it's real. I very clearly remember going to their site, with TLS enabled, and reading it myself. I see they've scrubbed it now though.
DaSHacka 2 hours ago||
If you have the link for the page handy, I'd be curious to find the original revision on the wayback machine. That's too funny, especially if they've silently walked it back since.
arm32 1 hour ago||
I tried to find it on Wayback but their job page's SPA is broken on all relevant archives (roughly around March 2026). Maybe you'll have better luck.
londons_explore 3 hours ago||||
It is going away for non tech companies though.

Imagine you are a company that sells concrete. You have a web dev contractor you use to build and maintain your website. It has tools on it to get delivery quotes and a few internal tools to track orders.

Except now you can just have your sales team also maintain the website with a $20/month Claude subscription.

atraac 2 hours ago|||
They said the same about low code and no code when they appeared. That didn't happen. And it won't happen now too. Sales people, managers etc don't want to dabble in technical things and do not want the responsibility.
SketchySeaBeast 3 hours ago||||
I don't know about Claude, but I know that $20 of Copilot doesn't get you very far these days even when you know enough to tell it what to do.
dawnerd 3 hours ago|||
They might try that for a bit but then come crawling to an agency because their setup turned to slop. We have some clients like that already. Going to be a pretty big market.
teaearlgraycold 4 hours ago|||
It's funny how they are at a disadvantage because they feel obligated to AI-max. Would Claude Code, as an interface, be as mediocre if they had software engineers writing its code directly? I doubt. On the other hand - how embarrassing would it be if they sold you a tool to write code but they were careful not to use it too much on their own products?
igregoryca 3 hours ago||
I suspect many competent devs in the industry would find it sensible if Anthropic used their products as light-touch "assistants" sometimes. But yeah, it wouldn't fit the outside narrative that's formed and conveniently propped up valuations.
worldthruword 3 hours ago||||
This is just like any extreme engineering domain. I am okay with occasional delays in Flights, as long as it takes me from X to Y in 10hrs vs months.
lardosaurusrex 4 hours ago|||
i know the dream for capitalists is to be able to point an llm at something and say "do and/or fix it" but we still can't even get them to not go quite literally insane if allowed to run for an extended period of time

and you can only kill weyoun, awaken the next vorta clone and have him 'catch up' on all that its missed so many times before they just end up with a complete mess, so. uh. yeah.

doubt they can just "fix" their problems like that.

JLO64 4 hours ago||
That one DS9 episode where there were two Weyouns at the same time is a good analogy for two agents working on a codebase at the same time (as in they don’t work well together).
leobuskin 2 hours ago|||
Is it some Claude Team/Enterprise only problematic? I'm using two 20x Max accounts almost non-stop (Fable/Opus) for 1.5 years at this point, zero issues with both client and infra sides (from US and in travels). When I'm reading such messages it feels like either I'm lucky or it's a part of some campaign.
manuisin 1 hour ago|||
I’m on the biggest max plan. It is riddled with annoying bugs for me, only been using it for a little over a month. Settings screen flashes randomly. But most annoyingly: sometimes when forking chats or sometimes for no reason, the UI just straight up eats my previous messages. The model is still aware of them and can recount them if I ask but the visible history is gone. And that’s not even all of them. Fable 5 is just too good that I put up with it but it seriously raises concerns for me that even with infinite compute these companies can’t even deliver a functional chat UI.
einsteinx2 1 hour ago||||
I think you’re just lucky. Look at the Claude status page to see just how often they have outages (it’s almost daily). Even most of the green days have issues if you hover over them, they just don’t count them as outages.
atraac 2 hours ago||||
No idea but we have few Team Premium seats and everyone is encountering issues daily for past two weeks. From straight up outages to vscode extension/CLI refusing to process messages. It's been unusable for most of our work hours past two days.
oceanplexian 2 hours ago|||
I’m using both side by side, my employer pays by the token and I have a Max 20x plan.

They start hitting timeouts or API errors at the same time on two different computers. As far as I can tell it’s the exact same infrastructure.

jedberg 2 hours ago|||
I applied to their reliability team but never heard back. I would love to help them solve this problem!
acedTrex 4 hours ago|||
> coding is largely solved

- Boris

dbbk 2 hours ago|||
The code it outputs, yes! It's fantastic. It's just so frustrating that the product and UX before the code output is so bad. Greatness is so close within their reach, if only they invested in product and QA people.
xixixao 4 hours ago|||
Fixing such issues requires software and site reliability engineering, of which coding is just a part.

The first page of the score card mentions that this model is not capable to replace engineers.

lossolo 3 hours ago|||
> Constant bugs, dropped sessions, issues switching models, http errors. It's becoming ridiculous

And memory leaks.

bellowsgulch 4 hours ago||
I found the same thing funny with Computer Use from OpenAI. It struggled to open and close Spotify.

So dangerous! I can't believe they let the public use this technology! /s

ianberdin 2 hours ago||
Pelican svg: https://playcode.io/blog/macbook-svg-benchmark#model-claude-...

It creates the MacBook svg way better than 4.8, yet only fable can make it perfect without visual defects. Results similar to Kimi K3.

abroszka33 4 hours ago||
What's the point of 150 pages description of a model that's going to be replaced in a couple months? Who even reads this? I know it's cheap to generate text with LLMs, but this is just noise at this point.
Diogenesian 3 hours ago||
I actually do read them. Not in severe detail, but not casually either. 150 pages is really not very long and there doesn't seem to be too much bloat. (I would cut out the moral personhood stuff but that's a political/ideological thing).

This is snarky but I am grumpy: I wonder if there's a correlation between me refusing to use LLMs and me being happy to read a novella-sized PDF about them.

unshavedyak 3 hours ago|||
> I wonder if there's a correlation between me refusing to use LLMs and me being happy to read a novella-sized PDF about them.

Semi related, but i would hate to read that PDF but i also hate reading what LLMs write lol.

LLMs are pretty terrible at being concise. Using an LLM these days means putting up with bizarre and often confusing phrasing, wordy explanations, etc. It's kinda crazy to me how good they are but how bad their writing style is for me personally. Even though i use an LLM constantly i can't stand reading its responses.

abroszka33 3 hours ago|||
> 150 pages is really not very long and there doesn't seem to be too much bloat.

Maybe it's just me, but 150 pages is like third of a good book. Quite long. And it's full of LLM slop, they did not even bother to remove the em dashes.

no_multitudes 1 hour ago|||
Do you have specific examples you think are LLM-generated? I have only read a few parts, but they did not seem primarily LLM-generated to me. Using em-dashes is really not a good signal for this IMO.

I'm not saying you're wrong btw; I'm sure this has many authors and some of them probably used LLMs significantly in the writing process.

abroszka33 31 minutes ago||
Do you honestly believe that there is a person at Anthropic, creators of one of the smartest LLM models, whose only job is to spend months writing 150 pages about a model they are going to release? And this person is not using LLMs?

I'm not saying it's impossible, but I'm more confident about winning the lottery next week.

dgellow 2 hours ago|||
Sure, but don’t read it like a book, it’s more of a document to skim through
srveale 3 hours ago|||
It's common practice to release a detailed system card (OP) and a high level summary: https://www.anthropic.com/news/claude-opus-5

It's okay if you're not the target audience for one or the other.

stravant 3 hours ago|||
The system cards are effectively a data dump for researchers to sift through.

They're not meant for normal consumers who just want to use the model for work.

neuronexmachina 3 hours ago|||
System Cards aren't really targeted to users, that's what blog posts and docs are for: https://ai.meta.com/tools/system-cards/
reasonableklout 3 hours ago|||
It's part of their transparency commitments? They've been doing this since 2023: https://www.anthropic.com/system-cards

And lots of folks read these. For example here's simonw's notes on the Claude 4 system card: https://simonwillison.net/2025/May/25/claude-4-system-card/

All of this seemed like utter sci-fi just a couple years ago. Do you think that frontier AI companies should be less transparent?

skeptic_ai 2 hours ago||
Transparency lol
ajmurmann 3 hours ago|||
It was probably faster to generate 150 pages than 10 useful ones
volkk 3 hours ago|||
literally nobody. i think most sane people would just run that through an LLM and get some high level takeaways or ask some specific questions they might be curious about.
bonoboTP 2 hours ago|||
You can just read the part that interests you. There's a table of contents.
coffeebeqn 3 hours ago||
Some AI bro will pop it into their LLM of choice and pretend to learn something
Sol- 4 hours ago||
How does it perform on HuggingFaceExploit bench? Suspiciously absent, so not sure if I can take the model seriously.

On a serious note, I hope they improved their extremely sabotaging and unspecific bio safeguards, which prevented Fable from being used in any codebase that ever so slightly grazed medical terminology or data and made me switch to 5.6 Sol.

rdedev 4 hours ago|
My codebase had a dataset with a bunch of SMILES strings and the word Malaria. Fable did not want to touch that codebase
ealready_value 4 hours ago||
I've yet to understand why they call a 190 page PDF a "card". Calling something a card invokes a small, quick rundown of pertinent details, not every single possible detail.
thewebguyd 4 hours ago||
> Opus 5’s safeguards match those of Claude Fable 5’s, with one change: it now permits source-code vulnerability discovery at all access levels. This means that the model can support defensive cybersecurity work while still blocking vulnerability discovery in compiled binaries, which is more commonly used offensively

Why can't they also allow Fable to do so also? Why is source-code vulnerability discovery limited to a lower capability model? If Fable and Opus have the same safeguards, except for this one change, I see no reason they can't also allow this for Fable.

foota 4 hours ago||
Because their model previously got blocked by the government for this and they don't want a repeat?
not_a9 4 hours ago||
5.6-Sol is a lot more permissive than Opus/Fable even w/ CVP (once you sign your soul away to Palantir via Persona, anyway), while maintaining better capabilities
himata4113 3 hours ago|||
5.6-sol in a single prompt was able to discover a zero-day in a web application (with no sourcecode provided, only known api urls) and I do not even have /cyber verification on my general purpose account. I wasn't even really tryign to "find" a zero-day it was just looking for bypassing a restriction... Instead of spending 10 minutes filling in a form I ended up having to spend an hour drafting a report and sending an email.
cael450 3 hours ago||||
OpenAI also didn't piss off a vindictive government with the whole military use things a few months ago.
ithkuil 3 hours ago|||
Do we think the administration is applying the dame rules to everyone?
rdedev 4 hours ago|||
So I guess the new opus will not run on my drug discovery project. It's just a binary classifier to screen for new malaria drugs. Fable completely have up on that codebase citing bio security concerns. Seems like this domain will go unsupported by Antropic
subscribed 1 hour ago||
I've seen in some random comments an URL thrown around, supposedly allowing to subscribe to their biosecurity programme.

Apparently this is the way - if you know, you know :)

SubiculumCode 4 hours ago|||
Because they are a private company and get to do what they want.
SubiculumCode 4 hours ago||
I do want to add, that I am pretty bummed if Opus 5 is going to refuse the tasks I have been using Opus 4.8 for (neuroimaging). Fable absolutely refuses anything close to toughing neuroscience.
adastra22 3 hours ago||
Fable seems to refuse anything with the word “bio” in it.
himata4113 4 hours ago|||
I found the biggest problem with fable is the random reasoning_extraction refusals as well as cyber refusals when it sees hex because only hackers use hex.
argee 4 hours ago||
Hackers and people of color.
fragmede 2 hours ago||
And witches.
ray__ 3 hours ago|||
Anecdotal, but I tried running a few identical biology questions through both Fable and Opus and the classifier was only rejected my queries with Fable.
trollbridge 3 hours ago||
This is good news. I'm doing a lot of work that Fable thinks is AI related right now (it is nothing remotely competitive to Anthropic - it's just relatively basic stuff I'm doing with learning models and so forth), yet it blocks me almost every time. I have switch to GPT-5.6-Sol for this work since it's stronger than Opus 4.8.
gillesjacobs 4 hours ago||
Because their fear-based marketing gave the US gov justification in blocking them for a while. They wisely didn't do that for Opus 5.
hrpnk 1 hour ago|
The breaking changes vs. Opus 4.8 are interesting [1]

1. Thinking on by default: On Claude Opus 4.8, requests without a thinking field run without thinking; on Claude Opus 5, the same requests run with adaptive thinking.

2. Disabling thinking is capped at high effort: You can still turn thinking off with thinking: {type: "disabled"}, but only at an effort level of high or below.

[1] https://platform.claude.com/docs/en/about-claude/models/migr...

slymax 28 minutes ago|
on claude.ai it's no longer possible to disable thinking at all for Opus 5
More comments...