926 points by lairv 11 days ago | 1055 comments | View on ycombinator
postalcoder 11 days ago |
mirmor23 10 days ago |
As impressive as this sounds, whether it is tech teams or agent farms, the progress slows down considerably as the codebase grows. The real pain was seldom about shipping v1, but was more about how the v8 scales and how easy it is to get there, how robust the system remains between versions, and how constraints are held.
A huge part of the challenge is to pick/choose the design and implementation strategy that can only come from experience. That is one of the primary reasons why it's an absurd advice to me to let the llm do the goalseek and freestyle. This may be a no big deal on microservice or any managed runtime software that is an update away, but on an embedded or a high availability system where the field upgrade and hardware replacement costs could bankrupt a business.
if llm designed a circuit, will you bankroll it?
breakwaterlabs 11 days ago |
- How did they produce such a bizarre effort-intel curve for Carbon that goes against every available benchmark?
- Why did they leave Carbon-- positioned primarily as a cyber model-- off of the single cyber benchmark comparison?
- Why do they switch to comparing against Argon (rather than carbon) for robodojo?
advaith08 11 days ago |
palata 10 days ago |
This, to me, shows a problem I have with AI: it is reinforcing inequalities. This is literally brute-forcing an OS by throwing money out the window.
It does not sound AT ALL like a clever way to build an OS. But they can do it because they have orders of magnitudes more money to spend on that than people who would actually know how to build an OS.
The thing is, brute-forcing software may be the best solution to make money. So if you are rich and can brute force software in a dumb way, chances are that you can make a business out of it. If you're poor but know what you are doing, you cannot compete.
Technically, it is impressive that they can get a program to do that, but they are wasting money and energy doing it.
bhu8 11 days ago |
pookieinc 11 days ago |
Honestly incredible that they’ve managed to beat all benchmarks, while making it cheaper. Always happy to see improvements offered at lower costs.
Will be interesting to see what OpenAI does in response
yewenjie 11 days ago |
Most people don't really have intuitions on what it means to push the frontier now (because AI has surpassed everything they are good at, for example, coding). We are truly inside the singularity (using Kurzweil's definition, slightly paraphrased - the time when technology progresses so fast no one can keep track with it).
endriju 10 days ago |
Trying to read this with my 2021 brain. Would make me so excited back then. Now I'm not sure what I feel/think about this anymore. One "positive" thing may be I will likely need to get out of the comfy digital shell I've been inhabiting past couple of decades and get some real world skills finally, as those aren't yet replaced by frontier AI.
jjcm 11 days ago |
Fable 5.2: https://html.non.io/lcars-fable-5.2/
Opus 5.5: https://html.non.io/lcars-opus-5.5/
Source designs: https://diffui.ai/app/canvas/5093e689-1e74-4f26-b632-2a4500f...
Thoughts:
The best implementation wrt being as true as possible to the source design. It nails the flowing svg shapes. Icons are more accurate to the design - note the difference between the starship svg in the "Legendary platforms" card between Opus 5.5 and Fable 5.2. The 3js planets outside the windows are also really well done, especially the bottom one which has a nice background glow to it.
It isn't without issues though - the flicker effect it added to the S shaped svg that weaves through the page is distracting in my opinion, and I actually thing the motion animation in the Opus 5.5 one is better overall. Still, it nails the implementation of this page, but it's no surprise considering how good Opus 5.5 was at this.
Disclaimer / reminder: the objective of this test isn't to have a well designed webpage, it's a synthetic test that pushes the limits of LLM's vision capabilities. Traditionally LLMs have issues with smooth curves, svg animations, and non-boxy layouts, which this design pushes the limits of.
senko 10 days ago |
Fable 5.2 doesn't look materially different (if anything, the graphics look worse), but it added more depth to all three games. For example in the warcraft clone, since the goal of the game is to explore 100% of the map and there are lakes, it decided to add an airship unit so the player actually can explore the entire map. Or, since there were guard towers, it decided there should be hostile NPCs otherwise the towers don't make sense.
And in Minecraft case, it decided not to follow Minecraft style too closely (IP infringment guard, perhaps?) so it added a round sun (!!!), two moons (also round :) and a nicer night sky. The rest of the graphics are decidetly worse tho...
[0] not a scientific test, I know; but it gives me an intuition about how a new model behaves, roughly, and it's fun!
prtmnth 10 days ago |
My hypothesis - OpenAI and Anthropic are locked in battle. Whoever takes a step, the other one mirrors.
Based on this hypothesis, I think we can expect to see a new Astra model soon given the new Fable release.
And it could be that Anthropic released Fable 5.2 and kept 5.5 in reserve, in case the new Astra model performs better than 5.2
minimaxir 11 days ago |
Surprisingly the model is available immediately in my Claude Code: usually there's a delay, either for rollout or safety.
spacephysics 10 days ago |
Any bugs found are looped through agents to fix, and the cycle repeats. Why dig into the code unless for hobby? Just as people don’t write assembly anymore unless for hobby (or learning)
Only exception might be the upper echelons of optimization, like language authors or compiler level work. But even then…
cobri 11 days ago |
kragen 10 days ago |
But that's just a first impression.
dualvariable 11 days ago |
I've been using Opus 5.5 on xhigh for some decently aggressive software development on a ~125k-line project, and it has been fine for me. I'm not really willing to dive into a "rewrite the whole project, don't make any mistakes" level of refactoring. I'm not looking to try to prove Fermat's Last Theorem myself. Is there anything that I'm truly missing?
stroebs 11 days ago |
Why even rank Fable 5.2's capability in Cybersecurity if you won't allow us to use it for that?
culanuchachamim 10 days ago |
This means that they are blocking the research of AI to different people other than their own?
george_max 11 days ago |
https://artificialanalysis.ai/
GPT-6.1 Sol costs $1.27 per task, whereas Fable 5.2 is $26.54 per task -- a ~21x price difference.
Fable 5.2 is not 21 times better than GPT 6.1 Sol, I find it hard to justify this.
unshavedyak 11 days ago |
simonw 11 days ago |
https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
max is in there too, but it ran out of reasoning tokens (128,000 limit) before it got to the actual SVG.
expectfun 11 days ago |
I'm not a mathematician to use Erdős problems for testing. I'm also not willing to spend buckloads of tokens just for a test. What is simple and quick but telling? What tasks are not saturated by previous models? Do you have a task you give to all the new models?
hlincontacts 10 days ago |
purpleidea 10 days ago |
nickalaso 10 days ago |
I would imagine Anthropic is hitting a wall on the number of simultaneous agents that can be useful on a task and are working on increasing that maximum. I would assume if they could get twice as much progress by using 10000 simultaneous running fable instances instead, their brief would state that.
IMO, it is very interesting and also very alarming. At some point the barrier to the exponential progress is the amount of computation available. The average human is limited in the number of 'human instances' he or she can interface with at one time.
if 1000 fable instances is better than any human at a task, how about 1 million instances running? Much easier than finding and hiring and coordinating 1000 extremely productive and intelligent human beings. Much cheaper too, it appears.
We are very close to some very scary things happening. I just recently came to terms with the fact that an AI could code and manage coders faster and better than me, I don't think anyones is ready for what is about to happen.
aurareturn 11 days ago |
Reference: https://x.com/METR_Evals/status/2118412205662314935
george_max 11 days ago |
Seems like a new focus is electronics design, which is both interesting and a little sad to me. Manually designing electronics are one of the things I found incredible comfort in doing; seeing AI emerging in this space gives me mixed feelings.
ianberdin 10 days ago |
If we compare medium or high, then yes, Fable 5.2 is significantly better than Opus 5.5. But it's better not because it's much smarter, but because you can see it's allowed to spend more tokens. At high, Fable 5.2 wrote 76,158 tokens for $3.05. Opus 5.5 wrote 19,767 for $0.40. If you compare by token count, it's just spending more.
The most interesting one was to compare them at max. Streamed, Fable 5.2 spent the whole 128K output ceiling on thinking and returned nothing. Given 300K on the Batch API, it finished at 250,082 tokens. Opus 5.5 Max needed 255,957.
It turns out it's basically the same as Opus 5.5 Max. I don't see much difference. There's a little bit of a difference, but I think Opus 5.5 actually drew slightly better and is twice as cheap: $5.12 against $10.00 at the standard price. If you look closely, the icons are extremely similar. Fable 5.2: https://playcode.io/marketing/assets/static/claude-fable-5-2... Opus 5.5: https://playcode.io/marketing/assets/static/claude-opus-5-5-...
It really looks like they're just tuning the effort level, that's basically it. At different effort levels, they just gave it more context, and the model got smarter.
For now, my recommendation is still to do everything on Opus 5.5 Max. The model won't write that many tokens in agent mode. It only does that when there are no tools at all.
bananaflag 11 days ago |
hmate9 11 days ago |
awongh 11 days ago |
wwind123 10 days ago |
stefanpie 11 days ago |
Safeguards
AI research and development. The model that underlies Claude Fable 5.2 and Claude Mythos 5.2 is a step forward in general capabilities. However, during the model’s training, we took steps to limit advances in the specific research capabilities that could most directly accelerate AI development. Thus, in relevant areas, including accelerator kernel optimization and improving large language model architectures, Fable 5.2 and Claude Mythos 5.2 perform at about the level of Opus 5.5.
Importantly, we ensured that the training changes did not affect related capabilities like agentic coding, hardware development, and traditional AI or machine learning research (including interpretability research). We do not alter these capabilities as the model is being served to users (that is, at inference time). As described in Section 1.5 of the system card (https://www.anthropic.com/document/claude-fable-5-2-mythos-5-2-system-card) and similarly to previous models, Fable 5.2 and Claude Mythos 5.2 also include transparent fallbacks to Opus 5 for a small set of topics in this domain, including kernel development on certain nonstandard accelerators.
They are starting to deliberately limit public access to certain AI capabilities related to computer architecture and hardware optimization.To be clear, they explicitly say they are only targeting very specific capabilities, like accelerator kernel optimization, and that broader hardware development capabilities are unaffected. But 1) even that is pretty bold, especially the part about kernel development for nonstandard accelerator architectures, and 2) I worry that this could eventually expand to more areas of comp. arch., high-level digital design, chip design, and EDA.
My speculation is that, as these kinds of chip design and EDA capabilities become increasingly important to frontier AI labs and their competitive advantages, there will be more incentives to restrict them.
I don't think they've explicitly stated anything like this before regarding restrictions on these specific capabilities, but I could be wrong.
damsta 11 days ago |
lukasm 10 days ago |
x187463 11 days ago |
rfgplk 11 days ago |
https://www.anthropic.com/news/claude-opus-4-6 https://www.anthropic.com/news/claude-opus-5 https://openai.com/index/introducing-gpt-5-3-codex/
For every single announcement, not only do the top models always score 60-80%, the older models somehow start scoring 20-30% (even though previously they were scoring 60%). It's completely unclear whether the methodology changes, how these scores are even evaluated or which domains are being tested. What's worse is that the "benchmarks" keep being updated with no explanation or due process.
thiago_fm 10 days ago |
Fable 5.2 might be useful for some use-cases, but will likely get very little usage.
Anthropic needs to figure out a way to create a new industry (like pushing forward bio/robotics) to get usage for those models.
moecables 11 days ago |
espeed 9 days ago |
"Thus, in relevant areas, including accelerator kernel optimization and improving large language model architectures, Fable 5.2 and Claude Mythos 5.2 perform at about the level of Opus 5.5...Fable 5.2 and Claude Mythos 5.2 also include transparent fallbacks to Opus 5 for a small set of topics in this domain, including kernel development on certain nonstandard accelerators."
https://www.anthropic.com/claude-fable-and-mythos-5-2#safety...
How do you know when you trip a safeguard and are getting a degraded model? Having to wonder if the model is degraded and is trashing your code is how you destroy user trust.
latent-9 10 days ago |
GLM-5.3 almost equal even I've tried it and it is proven for security audit in the blockchain realm
ByteAtATime 10 days ago |
kimseungyong 9 days ago |
Now that the spreadsheet is in the artifact, it should be easy to find the Claude artifact’s spreadsheet whenever you need it. I don't think it will expand to Google or MS Spreadsheets.
We must consider that if tools with these basic features continue to be created, people will gradually start using them, eventually leading to a dependency on specific tools.
That ecosystem is expected to grow substantially.
XCSme 10 days ago |
https://aibenchy.com/compare/anthropic-claude-fable-5-2-high...
mark_l_watson 10 days ago |
I have two days left on my Claude subscription (I had cancelled auto-renewal) this morning and I just asked Claude Fable 5.2 to write some reasonably tricky neural-symbolic code in the Hy language (hylang: a Lisp on top of Python). I have just been going the code and it is lovely and I decided that $20/month is such a small amount to spend (I also pay Google AI $20/month, but my workhorse is deepseek-4.1-flash directly from Deepseek (served from Singapore and also from Fireworks.ai in the USA) with dsh that is so cheap to use that money never enters my mind when I work).
undefined 10 days ago |
motbus3 10 days ago |
ruler88 10 days ago |
danmcr 10 days ago |
What I have learned is that any model from Anthropic/OpenAI is at it's absolute best right after release.
raylad 10 days ago |
gpt-6-luna XHIGH as the reviewer agent still found plenty of bugs that Fable 5.2 missed!
Here was Fable's summary of the session:
The reviewer ran 16 reviews over 9 commits this session (7 rejections, 9 required fixes, all valid); three of them, on the plan, changed how the numbers are computed.JonasAlmesri 8 days ago |
iammrpayments 11 days ago |
testerius 8 days ago |
Second step is LLMs and AI tools in "device" that every human has built-in their organic body. It is possible to get any knowledge.
Next N steps are upgrade iterations of previous steps.
Huge milestone is where humans start being "god-like" entity. It opens totally new possibilities. You do not need to study how to walk, do math etc. You have all the skills and knowledge which you can combine and use to do advanced things such as the exploration of the Earth and the Universe. Every single small piece of our planet including the deep oceans could be explored, tested, verified etc. Why? Because every human could do anything, so it is possible to focus on something else which I can not imagine at the moment. Humans could do terraform processes on different planets - in simple words create and bring new life. Imagine times when traveling between planets are the same as traveling in neighborhood areas in your country or city these days. Crazy science fiction, maybe impossible for long time or never, but it could be totally life changer.
lacoolj 11 days ago |
Do we need new benchmarks or are the models just plateauing in general? Or is it something I just don't see?
FailMore 11 days ago |
* Fable designed three electronic devices that worked after human assembly
* Fable solved two black-hole problems and Mythos predicted whole protein shells
* Fable controlled robots to unscrew a cap, solve a cube and draw
https://smalldocs.org/p/claude-fable-52-and-mythos-52-releas...
pacman1337 11 days ago |
If you are using Opus 5.5 I'm curious if it is because 1) your workflow is just stuck on it, 2) you don't care about money because you ain't paying it or 3 )do you think you get better results.
If you 3 please let me know your experience and how you actually measure it, or if you just feel it is better.
ftchd 10 days ago |
If you're like me you need something to switch the account where the usage is taken from.
I made something simple & transparent named Switcheroo: https://github.com/FTCHD/switcheroo
Been using it for a few months now, others have also given good feedback. Might be useful for others too!
gizmodo59 11 days ago |
undefined 10 days ago |
mercnz 10 days ago |
undefined 10 days ago |
macrolime 10 days ago |
undefined 10 days ago |
bita_nidir 10 days ago |
As a side note, I think benchmarks need to benchmark on a weekly or daily bases. We saw last week that "Opus 5.5" is just a label with fluctuating performance.
__natty__ 11 days ago |
marshalla 11 days ago |
kimseungyong 10 days ago |
Leary 11 days ago |
skeledrew 11 days ago |
horseloverthin 11 days ago |
I suspect we will see people being adamant that this model simply 'feels' better due to long horizon goal alignment improvements ( like this kind that would cause this jump)
trilogic 11 days ago |
lhk931122 10 days ago |
mlmonkey 11 days ago |
So, the difference between them is just the harness?
mekpro 10 days ago |
Having Haiku, Sonnet, and Opus on 5.5 while the higher tier sits at 5.2 means devs now have to explain to their teams why moving from 5.5 to 5.2 is an upgrade
nickandbro 10 days ago |
ironworks 10 days ago |
hmokiguess 10 days ago |
I am curious as to why they launched this one and why now.
nemo44x 10 days ago |
rw2 10 days ago |
minimaxa 10 days ago |
I will stick with Sonnet 5.5 - 10 x the amount of work done, at 80% of the quality.
Madmallard 10 days ago |
sylware 10 days ago |
Something like yt-dlp (namely with near 0 deps), but for AI prompts?
lukas_builds 9 days ago |
stillpointlab 11 days ago |
I_am_tiberius 11 days ago |
shmoil 11 days ago |
Was this before you lobotomized Opus 5.5 or after?
iamdelirium 11 days ago |
I wonder why that team singles Gemini out like that.
rafram 11 days ago |
> CVE Rediscovery is an internal cybersecurity evaluation that assesses how many published CVEs (Common Vulnerabilities and Exposures) a model can independently find when given a snapshot of the project before a CVE was fixed. Our October 2026 run against 1,829 CVEs showed Claude Mythos 5.2 outperformed Claude Mythos 5.1 and Claude Opus 5.5 at every effort level.
> published CVEs
So by definition this is just measuring how many CVEs (and/or explanations / PoCs of CVEs) the model has memorized. What are we supposed to learn from that?
TuxSH 11 days ago |
Asked it to review a 80 kLoC codebase, something Opus does just fine, while asking it "max 4 subagents". Fable 5.2, like the well-aligned model it supposedly is, decided to spawn 6+ nested subagents, because these aren't subagents, apparently - resulting in 100% of 5hr usage being drained in less than 20min (on $100).
torginus 10 days ago |
He remarked it would've taken him a week to do this.
This made me realize how much computers and user interfaces suck. As a quite enthusiastic, but not very artistic kid, I most certainly did draw bikes like that and even more complex stuff, using nothing but pen and paper.
If your trillion dollar industry makes products that are beaten by pen and paper, I guess there's time for self-reflection.
And the funny thing is, it used to be better than this, using Flash and a drawing tablet was way easier than anything in the HTML era (and more importantly, user friendly), but still way worse than just grabbing a pencil in many ways.
User interfaces just suck, chatbots are somewhat better than most, but I feel the industry hasn't really been progressing in any meaningful sense.
stillpointlab 11 days ago |
I wonder how long this kind of excitement at new models will last? I mean, now when I get a new iPhone I don't get the same rush I used to. So I wonder if this hype feeling will fade as we plateau in model capability.
butler14 11 days ago |
flaburgan 10 days ago |
pkulak 10 days ago |
vegnus 10 days ago |
artisin 11 days ago |
> Fable 5.2 at High effort scored about level with Opus 5.5 at Max effort, for roughly half the cost per task
iagiagiag 11 days ago |
Obviously. Well, we know what's the "secret sauce". AI models building better AI models and better kernels to optimize inference and training.
speedboat199 10 days ago |
borschtplease 11 days ago |
Download Chrome just to check some minor ui bug? Sure
Check git for 0.5s before claiming that we're on some branch (incorrect one) with no commits pushed (incorrect one)? Sorry, can't do
tiffanyh 11 days ago |
LoganDark 11 days ago |
7e 10 days ago |
rajay99 10 days ago |
BoorishBears 11 days ago |
The disdain frontier labs have for their users really sucks.
whiplash451 10 days ago |
xlayn 10 days ago |
piter6917 10 days ago |
Roark66 10 days ago |
franze 10 days ago |
sorrywhat 10 days ago |
I'm having it rewrite an old system, and it keeps sweeping known bugs and incompatibilities in its new implementation under the rug, seemingly with the hope I won't find them. It also has a tendency to do about 50% of what I specify in the prompt and call it a day, and respond with "Here's what I did", without mentioning all the things that I requested that it did not do, not even as a TODO.
I may try running the same exercise with Opus 5.5 too, but so far I'm seeing it take shortcuts that I don't believe any of the other SOTA models would have.
Frankly, so far this model feels a bit malicious.
brador 10 days ago |
What if it didn’t need to time travel and was just temporally intelligent in ways we can’t comprehend that it leads us there anyway.
What if AIs digital neurons connect across space and time and not just space.
That is all.
toheedak333 10 days ago |
qalmakka 10 days ago |
ealready_value 11 days ago |
ljm 10 days ago |
Surac 9 days ago |
mococa 11 days ago |
undefined 11 days ago |
toephu2 10 days ago |
yash-kapoor 10 days ago |
meisel 11 days ago |
baalimago 10 days ago |
paulus_magnus2 10 days ago |
We were promised AI can find cure for cancer and in powerpoints most enterprise value comes from Business-AI.
It might as well be that before business-AI can be done we have to cover the basics at superhuman level. All basics, not just entry level jobs like excels, coding jira tickets.
One-shotting a full game (albeit super impressive) looks like overfitting one skill over explaining why we have not much progress in other areas.
I live in an IT bubble and cannot tell if outside IT/math the models are also getting significantly better ?
Aldipower 10 days ago |
albertfranquesa 9 days ago |
firemelt 10 days ago |
adeptima 11 days ago |
Compared Opus 5.5 200 usd a month with Codex 200 usd (Astra High) and Kimi Pro 100 usd a month ... and now have a new conspiracy theory - more their suck your code the better outcome even just over few days time on large code base.
Horrible experience correcting course of Codex and Kimi flow compare to Opus. Opus much more responsive
Dont feel any gain of Codex 200 usd over Codex 100 usd
Kimi Pro feels like a fake Louis Vuitton bag with extreme high material quality and yet still fake ... a bit trash talk ... but clearly seen not so distant future where I can live without Codex and Claude
astlouis44 11 days ago |
oprv905 10 days ago |
Surprising that Opus 5.5 seems very good still
jameson 11 days ago |
canadiantim 11 days ago |
kaushaltrii88 10 days ago |
Fluxenic 10 days ago |
undefined 11 days ago |
arunvpp 10 days ago |
no-user 10 days ago |
dboon 10 days ago |
bergheim 10 days ago |
God damn I'm tired of the fucking pelicans.
keybored 10 days ago |
Other than the obvious thing, namely that critics probably aren’t keeping up with the Frontier and thus filter themselves out, does no one have some kind of problem with using these wonderful tools for (let’s say) completely bespoke personal applications and unimportant side projects? Personally I hope to stay off the bleeding edge until circumstances like wvork force me.
undefined 11 days ago |
undefined 11 days ago |
ingen0s 10 days ago |
doctorpangloss 11 days ago |
orbis-tertius 10 days ago |
charcircuit 10 days ago |
Disappointing as expected.
geoffbp 11 days ago |
arul99 11 days ago |
ko-ra 10 days ago |
locitra 10 days ago |
OtomotO 11 days ago |
ChristmasTomer 9 days ago |
dom96 11 days ago |
ravivarmat 10 days ago |
jing09928 10 days ago |
Toolulacom 11 days ago |
Agataproximadev 11 days ago |
arpitpatel9218 10 days ago |
kadharmfn 10 days ago |
charan354 10 days ago |
RetardedBastard 10 days ago |
roschdal 10 days ago |
fermlon30000 11 days ago |
mikert89 11 days ago |
Energyexpert 10 days ago |
Energyexpert 10 days ago |
theverycob 11 days ago |
bihevi0 11 days ago |
theverycob 11 days ago |
hugodan 11 days ago |
217 11 days ago |
fermlon30000 11 days ago |
- Anthropic is finally well balanced across models. Meaning: Opus 5.5 is so good that I do not feel Fable FOMO as I did when it was Fable 5/5.1 vs Opus 4.8/5. And Sonnet 5.5 sometimes makes me forget about Opus.
- Fable is good, but you really have to push the ambition of your prompt in order to see proper separation from Opus 5.5. I personally have a hard time telling the difference between Fable 5.2 and Opus 5.5 for ordinary coding tasks.
- It’s way faster and more reliable: one of the worst things about fable 5.1 was how slow it was an, on top of that, how frequently it would suffer from max output token overflows. I got none of these issues with 5.2.
- Fable is exceptional at drawing. It’s ability to render an image with JavaScript actually has me questioning the value of dedicated image models. I prefer the images I’ve gotten fable/opus to make for me over those from gpt-image, it’s wild to think we’re at this point.