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

Claude Fable 5.1 and Claude Mythos 5.1

anthropic.com

Post reply on HN
Print view View original

771–780 of 1001 posts

Re: Claude Fable 5.1 and Claude Mythos 5.1

#771

Post by simonw »

tukHelix wrote: post #740
simonw wrote: post #144

Pelicans for thinking effort low, medium, high and xhigh (that xhigh one is pretty good): https://tools.simonwillison.net/markdown-svg-renderer#url=ht... I'm still waiting for effort max to finish. EDIT: I fixed a bug in my tooling so it now records summarized reasoning traces - here's that max pelican, which is a significant improvement: https://tools.simonwillison.net/markdown-svg-renderer#url=ht... Took just under…

Maybe you could show a side-by-side comparison of pelican images. One image doesn't really make the improvement clear for someone like me. That would be a great help.

I really, really need to pull the trigger on a dedicated site for that. Maybe I'll have Fable 5.1 build it.

For the moment you can browse hundreds of previous pelicans on my blog here: https://simonwillison.net/tags/pelican-riding-a-bicycle/

Re: Claude Fable 5.1 and Claude Mythos 5.1

#772

Post by NL807 »

felixrieseberg wrote: post #69

(I work at Anthropic) Beyond all the benchmarks, I think Fable 5.1 is a big improvement in writing style. It sounds a lot less stereotypically like other Claude models, has (imho) a much more natural style, and responds to my style instructions more reliably. More work to be done (and we will!) but reading better prose makes me so much happier. Another point I expect not to get much attention until it all happens at…

Not sure if something like this is already on the table, but I would like to see Claude responses more in line with Simplified Technical English [1] by default. I find those writing styles a lot easier to read. This has been standardised as ASD-STE100 [2]. I've seen few people making SKILL.md files with that in mind, which works great, but having this by default without invoking the skill command would be better.

1. https://en.wikipedia.org/wiki/Simplified_Technical_English

2. https://asd-ste100.org/

Re: Claude Fable 5.1 and Claude Mythos 5.1

#773

Post by mmaunder »

mlaux wrote: post #46

Looks like all three breaking changes are patches for inadvertent chain of thought disclosure. Someone found out (don't have the tweet handy) that if you created a bogus "think_deeply" tool and then forced the model to use it, it would output what is believed to be its raw thinking there - I believe the first breaking change stops this. The second two are aimed at people getting Haiku to repeat thinking blocks from o…

Making clear the scale of distillation they’re combating.

Re: Claude Fable 5.1 and Claude Mythos 5.1

#774

Post by j45 »

velcrovan wrote: post #121
felixrieseberg wrote: post #69

(I work at Anthropic) Beyond all the benchmarks, I think Fable 5.1 is a big improvement in writing style. It sounds a lot less stereotypically like other Claude models, has (imho) a much more natural style, and responds to my style instructions more reliably. More work to be done (and we will!) but reading better prose makes me so much happier. Another point I expect not to get much attention until it all happens at…

I have a pet theory that the Opus prose style/smell we all have grown weary of is due at least in part to the models writing more for themselves and each other than for humans. They're packing lots of signal into fewer words and they don't care if it sounds cringe because it works better as glue in long-running tasks. I'm also thinking of the 2017 novel "Void Star" where AIs who operate everything have long since lef…

It could also be a balance between more words being less effort per.. token, etc.

Re: Claude Fable 5.1 and Claude Mythos 5.1

#775

Post by EugeneOZ »

mceachen wrote: post #391
Zigurd wrote: post #365

What I don't see in the comments: "I had a specific problem I couldn't solve with the previous version of this LLM. But the improvements in this version unlocked the solution for me." What I do see in the comments: subjective improvement in text generation, possibly lower cost, some optimism about code generation, but some skepticism too. I use coding agents. To me they are very useful. But what I spend on them isn't…

I had two sessions this morning that prior fable and sol sessions were stuck on, where iterations just resulted in _different_ bugs. (One kind of tricky fe layout problem, the other was a backend refactoring that was complicated by trying to aggregate a couple prior sessions that crashed). I summarized each into new fable 5.1 sessions, and both seem to have arrived at reasonable solutions that only need a few nits re…

So the issue was in large context size of these old sessions.

Re: Claude Fable 5.1 and Claude Mythos 5.1

#776

Post by slashdave »

jumploops wrote: post #70

> For example, in testing by the investment firm Millennium, Fable 5.1 found the cause of a rare crash on their internal systems that none of their engineers (or any other model) had been able to explain after several years of trying. Say what you will about LLM-generated code, but stories like this give me hope that software will never be as buggy as it once was.

If only!

Re: Claude Fable 5.1 and Claude Mythos 5.1

#777

Post by northzen »

alin23 wrote: post #479

My main gripe with LLMs is the cringe AI phrasings that they use in UI elements. Pompous things like "Your keys, supercharged" or weird yoda-speak stuff like "searches the app remembers" instead of just naming the thing "Learned searches" .. you know, proper GUI copy like it was done for the past decades. I jumped when I saw a mention about "writing style improvements" so I gave it a try on a recent feature in rcmd […

It's watermarking.

This style provides a high entropy basis distribution, so they can from a bigger pool to pick from and phrases to watermark the sentence.

You have much bigger variation of this idiotic phrases and words, which states a simple fact in that sophisticated and twisted manner.

Re: Claude Fable 5.1 and Claude Mythos 5.1

#778

Post by qingcharles »

EliasWatson wrote: post #308

To be honest, these frontier model releases have become boring for me. Opus 4.8 was already good enough for most of my use cases. I don't have any projects right now that I would use Fable for instead of Opus. So when I see announcements like this I just think "that's cool I guess" and then go back to using weaker/cheaper models. What's far more exciting right now is models like DeepSeek V4 Flash and GLM 5.3 Flash. T…

I have a project that could use Fable, that fails horribly on Opus, Sol, Grok, Gemini and GLM etc, but I just can't afford to use it on something so frivolous. (Assuming Fable could complete it)

Re: Claude Fable 5.1 and Claude Mythos 5.1

#779

Post by solenoid0937 »

johnsmith1840 wrote: post #398
GodelNumbering wrote: post #139

The price reduction comes from the cache read pricing falling from $1/M to $0.25/M, which means that Fable 5.1 now costs half of Opus's cache read costs ($0.5/M). This gives a lot of credit to the theory that Anthropic did not get much bite on Fable at its original pricing, which in turn likely places a ceiling on LLM pricing in general. Interestingly also, if you take away terminal-Bench-Science 0.1 results, it is h…

I'm a heavy user and fable is great the #1 reason I stopped using it was the horrible safegaurd filter. I found sol close enough in capability and have only been blocked when my request was an obvious offensive cyber work. Fable blocked me on almost everything. Optimizing a OS build? -> block Securing a container -> block 60% is nowhere near enough for that safegaurd system. This just means I am going to be blocked h…

The safety filter is simply no longer an issue.

Re: Claude Fable 5.1 and Claude Mythos 5.1

#780

Post by z3dd »

rxyz wrote: post #346
GodelNumbering wrote: post #139

The price reduction comes from the cache read pricing falling from $1/M to $0.25/M, which means that Fable 5.1 now costs half of Opus's cache read costs ($0.5/M). This gives a lot of credit to the theory that Anthropic did not get much bite on Fable at its original pricing, which in turn likely places a ceiling on LLM pricing in general. Interestingly also, if you take away terminal-Bench-Science 0.1 results, it is h…

Anthropic did not get much bite because they don’t offer zero data retention with fable

This is exactly the reason why fable is blocked at the company I work at.
Post reply on HN