Hopefully the output from vanilla 5.5 is as good as they claim. I’ll try out later tonight.
It's just a standard hero image + text for me, with no scrolling effects.
I agree that it's sort of stupid, not a fan.
Sounds like they noticed the complaints. I'm curious to see what LLM-isms this one may have.
I don't mean to pick on this comment in particular. The majority of my work day is now spent reading AI generated text, and I look at HN (too much!) because I want to read human commentary. Humans pretending to be obnoxious AI on repeat is net negative to say the least.
Less companies involved means less pressure to go fast.
Is the Xbox 360 (Xbox 2) vs PS3 debacle all over again.
It was odd at the time, yes, but no one really minded it truly. Heck, Xbox “ONE” was a lot more of a fiasco/debacle than “360”—but there’s no parallels to be drawn with “ONE” here.
I see what you’re trying to get at with this comparison, but a “debacle” it ain’t.
We can't test it properly because it knows it's being tested.
They write that at the top, but then on benchmarks, it beats literally every other model, including Fable and Astra?
Will be interesting to see how people's opinions of it line up IRL, but so far I've loved Fable so hopefully will love this one too
Might have to use my $20 Claude sub some more. I was moving away from it to a $100 OpenAI one to avoid the Claudese and poor token efficiency of Opus 5, given that I couldn't use Fable 5.1 with my tier, but this is worth trying out.
> Input and output tokens are $4 and $20 per million, 20% less than Opus 5. Cache reads (which make up the majority of agentic and coding work costs) are $0.20 per million tokens, 60% less than Opus 5. Opus 5.5 also generates output more than 30% faster than Opus 5.
Better than Fable, cheaper than even the last Opus. I use Opus as my main driver so this is very exciting!
Maybe Anthropic finally felt the pressure from MiMo, DeepSeek, GLM Flash and Luna.
God I hope so
I’ll be going about my day, have a random idea, launch a microvm on exe.dev with a prompt of my idea, and get a working thing a few minutes later.
I don’t know how much better a model would have to be to get me to move off OpenAI at this point, but doing just a little bit better in terminal bench 4 isn’t it. It would have to be a difference in kind, like opening up the harness restrictions, or privacy guarantees (comparable to offline models).
Edit to address questions below:
ChatGPT supports oauth login.
Exe.dev has it built in. IIRC, pi also has it built in via /login.
Terminal-Bench 4.0 - Stanford & Laude Institute (with funding from all of the AI companies)
FrontierCode v1.1 - Cognition
CursorBench - Cursor (now SolarBoringSpaceXAI I believe)
GDPVal-AA - Artificial Analysis
AutomationBench - Zapier
Humanity's Last Exam - CAIS and Scale AI
Terminal-Bench-Science - Stanford, Laude, Ai2, Allen Institute
OSWOrld - XLANG Lab @ the University of Hong Kong
Chartography - Surge AI
> On our benchmarks, Claude Opus 5.5 leads in agentic coding, computer use, and knowledge work. That said, at these levels of capability we’ve found that benchmark margins have become a less reliable guide to real-world differences. In our own use, the gap between Opus 5.5 and Claude Fable 5.1 is narrower than these scores suggest.
In general, "benchmark margins have become a less reliable guide to real-world differences" sounds like a big problem. It was certainly the biggest problem with the previous generation of Claude models for a different reason, because the non-code output was nonsensical, and that is not being benchmarked at the moment. But I'm not sure what to make of this admission.
In my experience Opus 5 is the worst of all possible worlds, it's dumb and headstrong. It just runs away with tasks you didn't ask it to do, is reckless, and basically is unusable in my experience.
Not sure why but my guess is that this will be that bust worse. Happy to be proven wrong.
I've really gone in the opposite direction: having a dumber model orchestrate. In my case, it's usually a Luna orchestrator spawning Sol/Astra subagents to do the "big brain" work of planning and reviewing.
Reason I went with "dumb orchestrator" was just to save tokens. Having Opus/Sol (let alone Fable/Astra) orchestrate was burning tokens like crazy for me even when much of the gruntwork was being done by Luna/Sonnet/Haiku subagents. (Luna is also really good, like way better than Sonnet...) Perhaps it was a skill issue on my end though, maybe I wasn't just managing context properly.
> Reset for free: Get extra wiggle room to explore Opus 5.5. Expires Oct 22.
Prices per 1M tokens Claude Opus 5.5 Claude Opus 5
Cache reads $0.20 $0.50
Input tokens $4 $5
Output tokens $20 $25
Cache writes $5 $6.25
If you are forced to reduce price despite raising capabilities, that certainly tells something about the market, and potentially about Anthropic future profitability too, since this model is their biggest topline contributor
For long running tasks it is. That's what made Deepseek so cheap.
Holy shit! Its happening!
Now if we can the AI to understand this *implicitly* so that it doesn't need to be stated upfront, we might be able to undo years of "premature optimization is the root of all evil".
Ah, they're spreading their limits to all their models it seems. Definitely not a good thing long term in my opinion.
The real answer is local instantiations where you don’t have to worry about poorly tuned guardrails screwing you over while you try to work.
Until eventually the Chinese models get good enough/the strategic balance shifts and they start locking everything behind closed weights the same way the US companies are doing.
Infomercial at its best.
No wonder we are hammered with ai announcements.
I could see them accomplishing it and seeing gains like this in roughly the correct timeframe, and when I heard about that development I assumed the frontiers would probably jump on it.
How it works: https://miraflow.ai/blog/deepseek-v4-1-flash-causal-encoder-...
Interesting how the very first line is used to remind the reader of their call to pace the frontier just last week, and everything else after that line is to demonstrate with very specific numbers how they absolutely are not pacing.
doesn't sound like a razor at all
Anthropic has used "in the near future" for Mythos-class models too, but CVP is still Opus 5 only.
Why even have the program designed for trusted access to cyber capabilities if you're not providing access to cyber capable models via the program?
Nice. I was starting to think Haiku was going to be abandoned.
Great so good luck using this for any low-level embedded or operating system development (unless you really, really like Opus 4.8 and want to be greeted by its familiar face after a few minutes of work!)
Nice. I was starting to think that Haiku got abandoned.
I would maybe use Haiku 5.5 for highly parallel workflows like checking in on MRs or scanning my entire codebase.
Maybe its a bit tiresome to read another comment of the form "what about your large scale distillation attack on the Internet", but this statement really just pisses me off. How very insincere in the most aggravating way.
Thank you.
All of this starts to feel more like a drug dealer selling their newest stuff.
In two weeks we probaly get Fable 5.2 with “groundbreaking” improvements, then Astra x+1 etc and then the cycle starts again.
And on the way I always have to check my tooling and need to adjust things to get max results.
Benchmarks often don't survive contact with reality.
Instead of instilling confidence, it was overwhelming. Not sure if I'm the only one.
HN is a bubble that's mostly out of touch with what regular people use or care about.
In 2007, HN was convinced that nobody uses Microsoft products. In 2016, it was that Facebook doesn't have any real users and is dying. In 2026, it seems like nobody cares about AI safety and everybody wants to run local models.
"Early testers found its writing clearer and easier to follow, which addresses some of the common feedback we heard about Opus 5"
and
"We’ve made major improvements to the way Opus 5.5 writes and communicates, one of the most common areas of feedback we heard about Opus 5."
and
"In our own use, this has made Opus 5.5’s work easier to follow and check—which is a safety benefit as well as a practical one."
I realize it is corporate communications but "most common areas of feedback" and is a bit sterile. If the company wants authenticity and trust its easy to say that they found it hard to follow. And that it did not meet a quality bar they generally expect from their releases.
If this is not true, that it Opus 5 output was generally acceptable and we might see something like that again, that is an important consideration for potential customers or investors.
Such a negative tone they put on this. Distillation is amazing, because it means anthropic and openai fail to keep a monopoly. Who even are they who claim it's unethical? If it is truly unethical, then so is the mass data scraping they do on my personal website on a regular basis (without my consent), and all the unauthorized use of content produced by authors, blog writers, wikipedia contributors, and creators everywhere. If it is truly unethical, then anthropic, openai, meta, google... all these companies should have deleted their LLMs long ago. This wording disgusts me.
Heck, it would be amazing if we had more models without guardrails - some of the models that are produced via heretic[1] are actually quite nice to use - in particular, I've enjoyed investigating Chinese censorship by interacting with an abliterated model of Qwen3.8-27b. If security is really a concern, then secure your systems - don't attempt to dumb-down the tools we use. If someone breaks your window, then they are responsible, not the hammer they use to do so.
I tried Opus 5 and Astra.
A bit confusing, otherwise I would assume this is a complete replacement for Fable across the board??
Fable feels a bit dated after using Astra. Progress is nice (just like Grok 4.7) but I'm looking forward to the next big release as I can't fully commit to Astra atm.
I was accepted into the CVP a little while ago. Does this mean I'll need to apply again?
Wdym Opus 5.5 scores 14.7% higher than GPT Astra for Terminal Bench 4.0?
How would this alleged difference (most likely bs) actually show up in reality?
GPT Astra was literally the best model in the world by a margin until 1 hour ago or so.
>> On our benchmarks, Claude Opus 5.5 leads in agentic coding, computer use, and knowledge work. That said, at these levels of capability we’ve found that benchmark margins have become a less reliable guide to real-world differences. In our own use, the gap between Opus 5.5 and Claude Fable 5.1 is narrower than these scores suggest.
I think this is what I'm most interested in. I mostly moved to Astra because I just can't work all day with the Claude Opus 5/Fable writing style. I've don't think Astra is a better model, but it's the first OpenAI one that seemed good enough for me. Definitely keen to try Opus 5.5 and see if this claim is real.
Everyone that doesn't gets served some animated bullshit.
For a marketing page, it’s not the worst UX I’ve seen, but still slightly annoying.
and it's not about the verboseness (even though it obviously contributes to the fatigue and loss of focus), I swear the vocabulary of the llms change working on the same task on the same codebase significantly.
I wonder if there are studies around this.
https://openai.com/index/where-the-goblins-came-from/
Small quirks can quickly add up in posttraining if not caught. Although TBH with how obvious Claude language is, I do feel like this is something Anthropic probably noticed and just assumed people would not care about. Now that people have obviously cared, they're probably actively looking to alleviate it
I wouldn’t be surprised if Opus 5 was trained on content written by other LLMs
> The Vercel target is hard-coded. That's common and not wrong, but it's opaque; nobody reading this later will know which Vercel project it belongs to, and if the project is recreated the target changes silently. A comment or a named variable would help.
> Pointing a DNS name at Vercel is only half the job. The domain also has to be added to the project in Vercel's dashboard, otherwise requests will arrive and Vercel will reject them. That step lives outside this code, so it's easy to forget.
> Finally, [CENSORED] existing only in production is slightly odd on the face of it. It may be perfectly deliberate (perhaps a single shared testing tool that only needs one public address), but if you're reviewing this rather than just reading it, that's worth confirming.
It has the same annoying cadence and writing style with slightly less prominent claudisms.
* Consider leaving a comment about the hard-coded Vercel target. It's not clear where does it come from.
* [This is just a bullshit point, because the domain is not "added to" Vercel, it's provided by Vercel]
* Are you sure that [CENSORED] is prod-only? The name suggests otherwise. [also, what "if you're reviewing this rather than just reading it" even means?]
This is news to me. Excited to try it out! Thanks.
Can you give more details here? This sounds intriguing.
"Better" in every sense of the benchmarks and absolutely horrible results in my day-to-day work.
The verbosity, goal post moving, tendency to leave work unfinished, over focusing on unrealistic root causes when debugging, etc... etc...
It was the first time I actually pinned my models back because I just could not work with 5 for the price and performance it gave me. Hoping 5.5 is better this time around....
"benchmark margins have become a less
reliable guide to real-world differences"
sounds like a big problem.
1. Real-world use cases typically involve big, hairy, crufty, tech debt laden codebases and benchmarks do not.
2. AFAIK "success" in a benchmark essentially boils down to "do the tests pass and do we get the right result?" which is something the LLMs have been achieving with ease for a while, except maybe for uber-challenging coding tasks that would be outliers in just about any workplace. Whereas real-world software engineering is usually just a bunch of CRUD... and "success" involves harder to measure dimensions like "maintainability" and "did you overengineer this?" and "how did you cope with a bunch of vague and maybe contradictory business requirements?"
Having said all of that, I have never ever looked inside any of these benchmarks. I'm putting my guesses out here strictly in the tradition of "the quickest way to learn about something is to be wrong about it on the internet."
If 5.5 is any better, I might try to do agentic-assisted development instead of just telling fable to delegate
It does work out to be a similar cost per task though
The safeguards really don't work well for a lot of long-running tasks on old code bases. A lot of my workloads last days to weeks and the single biggest risk to the workflow is random safeguards.
What insights do you have? Because the blog/benchmarks don't imply any "ish" ... this is crushing Fable across the board. The only nuance to this is how the employees are saying what you're saying: "similar performance" but again what does this mean vs what is presented to us?
that, they fully intend to 'pace'. hiring accenture is a good sign they need some justification theatre and a fall guy for this decision.
I'm on the fence about calling out AI-isms but I think it's definitely worthwhile to call out ones that actually don't make sense.
They’re limiting frontier model development speed. Others are too. Pacing is the only word here to criticize, and I think it’s fine given the limiting of speed but also increased oversight. I’m not saying they’re fully doing this, but the term is fine.
Do you have a better proposed phrase?
throwaway2027•51m ago
handfuloflight•50m ago
cronin101•50m ago
staticman2•47m ago
rich_sasha•46m ago
cmrdporcupine•36m ago
danw1979•30m ago
aoeusnth1•36m ago
fghorow•35m ago
esafak•14m ago
ThouYS•35m ago
danw1979•32m ago
hmokiguess•31m ago
RGS1811•24m ago
carlos-menezes•22m ago
sailfast•21m ago
lgessler•5m ago
The outage is a real concern, and the issue is worse than we realized. Requests to Claude Mythos 5.1, Claude Fable 5.1, and Claude Opus 5 encountered elevated error rates. These models are load bearing for many developers, and that is what makes this outage really bite.