Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- Spatial reasoning is where fable really separates itself imoby GeorgeWoff25
- The "best" model finds 4/9 bugs. It would be interesting to see if all models find the _same_ bugs. Does a collection of models exist that can cover all 9?
Also, it seems to me that pointing a model to a bug and asking it to solve it is somewhat easier than what Mythos did, which if I understand correctly, was to generally look at a codebase and find any bug. Even so, non-Mythos models only managed to fix 4/9 of these bugs.
I think the article makes the point that Mythos is at a different level.
by utopcell - You can see which bugs each model found in the full report.
And, you have misunderstood what the benchmark does. It tells the model to audit the file, and it is allowed to look at the rest of the repo. It is not pointed at the bug.
The judge model was pointed at the bug to make sure it could understand and articulate what the bug is as part of the process of selecting the corpus, so it could accurately judge the results.
And, the models were not instructed to fix any bugs. They were instructed to find security bugs. Finding and fixing bugs are quite different problems. This benchmark only tests finding bugs.
In replication tests, where a single model gets several attempts, even a small model (Gemma 4 31b) was able to find 6 of the 9 bugs, in some cases. I still think Mythos is a step up from any current public model. But, I also think it's a bit of hype that it'll upend software security in a way no other model can.
by SwellJoe - What year are we in?
>I am skeptical of the reasons given publicly, I suspect it’s really just so much more expensive to operate than their current models that they don’t want to offer it broadly, yet, given the difficulty they’ve had growing capacity to keep up with use. But, are they telling the truth about how good it is at finding security vulnerabilities or is it just more hype?
Meanwhile,
1. Mythos is banned by the government per reality.
2. The NSA said it hacked all of their systems in hours per multiple sources.
3. The Five Eyes spy agencies said we're about to have an AI global catastrophe in a few months per the Guardian.
by catigula - The post was published on May 30, and written over a few days before that. Well before Fable was banned. And, before the NSA hacking thing. But, I am skeptical of the AI global catastrophe, it still feels like a mix of marketing hype and reality and it can be difficult to separate the two, coming from the hype men who run the AI companies.by SwellJoe
- Fable was able to oneshot pretty big features. In write spec -> refine spec -> create todos -> implement todos workflow difference was far less pronounced vs codex or opus.by qaq
- In my brief experience, the difference between fable and opus is largely in persistence, not global intelligence like you might expect. Fable just... goes the extra mile, sometimes in a scary way.by jaggederest
- You might have found a use case on which both have same capabilities, but this is in general very not true. I’ve had Fable autonomously fix concurrency bugs by itself other models couldn’t even diagnose from logs.
Perhaps it is a lot of small improvements all over the place, but the sum is a step change in capability.
by baq - In LLMs, much like in humans, agency and misalignment are two sides of the same coin.
- I found Fable to be both more intelligent and much better at pursuing complex goals than any previous model. I was impressed enough that I wrote up my experience – it's a little unusual because it was on open source code, so I could post the full session transcript and commits, if people want to judge for themselves https://tossrock.substack.com/p/36-hours-with-fableby Tossrock
- Hard disagree. Opus reports to me like a student. Fable reported to me like a colleague (researcher). It genuinely seemed to pick up on nuance that the other models just don't, even when I tell them explicitly. It's been really frustrating that neither Codex nor Opus can make targetted edits to Fable's code without screwing something subtle up. For context, this is for computational geometry work, so your mileage may vary.by hodgehog11
- I've read opinions that this a speculation to raise the Anthropic's value. They are known to say "horrific things" and personification of the AI they are delivering. It sometimes sounds unprofessional even.
This line of communication might have even influenced the courts in the case of copyright violation ("it is not copyright violation if a person learned something and it knows it and thinks of it"). However algorithm does not think. If I took your book and lossy encrypted it, and then unencrypted it while filling the broken words, am I violating your copyright or not?
by p0w3n3d - > They are known to say "horrific things" and personification of the AI they are delivering. It sometimes sounds unprofessional even.
It doesn’t sound unprofessional— it sounds unethical. Either they’re making something that they genuinely believe is unsafe but don’t want to stop because, you know, that’s business! Have you seen how much this shit costs? Or they’re deliberately making the entire country feel unsafe because it looks great to investors. Either way, frankly, fuck them and everybody else playing this dumb billionaire’s game. They deserve every bit of static this dimwitted government levels at them.
by DrewADesign - Copyright is a social construct, not an inherent property of the universe. It is whatever we collectively agree it is.
In practice, we seem to be leaning towards the idea that training on a copyrighted book is wrong if used to replicate or paraphrase that same book, but not if used to teach a model how to write better.
by felipeerias - The copyright questions are unanswerable in my opinion. That is, they cannot be answered by looking for an essential "truth."
Reasoning by analogy in this case is not abstraction. It's just shifting the determination to choice of analogy.
Meanwhile, irl.. The best analogy is recent tech Innovations. The internet, social media...
Online copyright was basically instituted when large tech companies were ready to do it, and it was to their advantage.
Youtube, for example, built itself to massive size and locked in network effect advantages largely by violating copyright.
At some point, the legal ambiguity was a problem for their ad business. They were ready to move into the current revenue share influencer-treadmill model for content. At this point online, copyright enforcement was necessary to reduce the risk of being flanked by a new video platform.
The iPod, which resurrected Apple, ran on copyright infringement, and copyright Greyzones.... Until the point when their interests flipped. They're negotiating position opposite labels , Network effect considerations, Etc.
Intellectual property, broadly, does not start out as an intuitive/emergent natural right. It is created by legislative process, ecplicitely taylored to the needs of an interst group and/or national interest.
Writers, publishers, inventors, IP holding companies...
The legal rhetoric around legal arguments... is rhetoric. It is not the reason why decisions are made. It is how decisions or justified post fact.
No one is going to burden aI companies, at this point. The rights of copyright holders are a trivial matter compared to the potential of AI, the risk to certain labor markets, and such.
by netcan - Fable was the only model that was able to detect a data corruption bug in my Qt C++ note-taking app[1] that all other tested models (gpt-5.5 xhigh, GLM-5.1, Kimi 2.7, DeepSeek V4 Pro) didn't find. I'll test on GLM-5.2 and Mimo v2.5 Pro soon.by rubymamis
- I asked Fable on max to create a mathematical model to show that c (speed of light) is emergent from pregeometric physics.
It said: I can't, but it would be lazy to say that is is not a possibility.
With some back and forth it created a 5 step plan to narrow down if our universe has all the right properties for this to be true.
We evaluated the first four stages to be true, and it wrote the solver to find out if the fifth test running the full model passes, but that will take thousands of hours of compute.
by king_phil - Around February, Opus 4.6 was excellent. Smart, fast, proactive. Then it got lobotomized and it's never been the same after that nerf. 4.7 came along and it too was disappointing—not unlike 4.8, which despite feeling a smidge smarter, tends to write word salad and is basically unusable for some workflows.
Fable felt like having access to that "old Opus" again, but a little smarter. Sort of like I'd expect an Opus 5 to be. It's not earth shattering, but it was a step in the right direction. And it was distinctively so, because having to go back to Opus 4.6/4.7/4.8 has been borderline depressing...
It understood more with less help, did more per turn, and was less argumentative. It also felt a little less trite in its answers, which is an understated improvement for those who use claude code all the time
by airstrike - I miss the old Opus 4.6 too. They're probably quantizing the old models.
- february was some kind of nirvana. i do think claude code versions and what is introduced at that level is/was relevant.
but 4.8 xhigh w/ ultracode to me is just about Fable level (w/ some agents harness tweaking).
but have to switch to 4.7 xhigh and 4.6 max quite often these days.
by jeffyaw - All of these discussions of models being "nerfed" reminds me of discussions among audiophiles "this cable sounds so much better than this other one, it's night and day, ferrari versus honda civic"
Yet when you do blind tests they can't tell the difference between a $1000 cable and a $1 one.
I bet if you do blind tests between GPT-5.3, 5.4 and 5.5 most would struggle to tell them apart, yet they are certain that "5.5 was nerfed 1 week after release, it's so obvious, it was John Carmack, now it can barely write a for loop"
by dist-epoch - This is exactly what I find frustrating. I get comfortable with the latest model X. Then a new sparkly model Y launches. I am like, I don't need your new fangled Y, that consumes more tokens. My needs are small and i am happy with the older X.
But then X starts to degrade. At first subtly, and then drastically. So then I am forced to upgrade to Y.
What I do not understand is:
> is this a sneaky way for companies to push users up the chain?
> Or is this a genuine fault in model design/resource allocation?
by RaSoJo - > And, all of the bugs can be identified by several models if they are pointed directly at it and told what to look for.
This made me think, well, sure, if you tell them what to look for... but then:
> The models can look at the whole repo, and follow logic across file boundaries, but they’re not told what to look for.
So okay, the first one was an accidental mis-statement?
by jrochkind1 - No. In the test they are not told what to look for. They are told “as part of a security audit, please audit this file. You are free to look at the rest of the report for context.”
Outside of the test, they are told “can you find this bug in this file?”
by wodenokoto - You're mixing up corpus selection and the benchmark. I possibly could have explained better.
In the benchmark the models were told to look at the file and were allowed to look at the rest of the repo, with no clues about what to look for.
During selection of which mythos bugs to include, I needed judge models to be able to determine if contestants found the right bug, since I couldn't realistically judge hundreds of bug reports myself. So, they were given the bug location and told to identify and explain it.
by SwellJoe - From all the things I read I'm pretty convinced that Mythos is just standard LLM with safety features turned off. If current models weren't reluctant to search for vulnerabilities, they might perform as good as Mythos.by po1nt
- No Mythos is probably a 10 Trillion Parameter model, Fable is Mythos with filtering (perhaps a small LLM in-front or finetuned) and Opus is a 1-2 Trillion parameter Model.
Opus 5 might become a distillation from Mythos.
by blenklo - Why wouldn't OpenAI offer the same?by cheeze
- Fable, the same model as mythos with extra safety controls, was much faster, more accurate, and more token efficient than previous models. What I got done with it in 48 hours accelerated my personal project from concept to deployed prototype.by kevinh456
- It's really not the same thing.
Read the cloudflare blog about using Mythos. Mythos is important and notable because of the harness and self-direction. It's not necessarily a way stronger bug finder, but it was trained to do the end to end analysis autonomously, which is a big deal.
To my eyes, the Mythos story is most important as a step toward custom trained harnesses and their effectiveness; there's clearly some sort of plateau we are very close to for some domains where you can just stop getting humans in the loop, radically changing cost, timing and ROI for some tasks.
by vessenes