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

Claude Fable 5

anthropic.com

Post reply on HN
Print view View original

731–740 of 1001 posts

Re: Claude Fable 5

#731

Post by 00deadbeef »

coreylane wrote: post #524

I dont get why Opus 4.7, 4.8, and now Fable all stopped supporting structured outputs? Does no one else care about that? I find it incredibly useful to reliably pass LLM output directly to other APIs/libraries

They do

https://platform.claude.com/docs/en/build-with-claude/struct...

> Structured outputs are generally available on the Claude API for Claude Opus 4.8, Claude Mythos Preview, Claude Opus 4.7, Claude Opus 4.6, Claude Sonnet 4.6, Claude Sonnet 4.5, Claude Opus 4.5, and Claude Haiku 4.5

Re: Claude Fable 5

#733

Post by 00deadbeef »

jeffhwang wrote: post #679

Is anyone else confounded by this naming scheme? I can see from the article's first two footnotes that Mythos is supposed to be a tier above the standard Haiku/Sonnet/Opus sequence. Ok that's fine since we learned about Mythos and Project Glasswing earlier this year. But now there is Fable--and why "Fable 5" even though this is a first launch? How is it related to Opus 4.8, Sonnet 4.6, Haiku 4.5, etc??

The first number is which generation of their LLMs it belongs to.

Fable is the first model in the 5th generation.

The second number is an incremental release, not a generational leap forward.

Re: Claude Fable 5

#734

Post by josephg »

kansface wrote: post #715
simonw wrote: post #534

I've spent enough time with this now in Claude Code (and Claude.ai and Claude Code for web) to have an opinion on Fable 5: it's a beast. I'm throwing some VERY difficult problems at at - things I've been dragging my heels on for months - and it's crunching through them very happily. One that I'm willing to share (albeit from just a week ago) - I built a Python library last week that bundles MicroPython compiled to WA…

Yes, exactly this. If I didn't care about price at all, I'd exclusively use this model. It functions more like an actual engineer. I'm in the midst of a DB migration, and eg 5.5 continually suggests stuff like "use DB X instead of DB Y for task Z because its 30% faster" which is an impossibility of reality, given we are migrating DBs. Fable jumped in, reduced allocs by literally 46x, found multiple bugs 4.8 and 5.5 c…

I'm having the same experience. I'm in the process of implementing a new CRDT for realtime collaborative editing. There just aren't a lot of implementations of CRDTs kicking around online for opus or any of the other models to have good design instincts.

Fable is doing - so far - a great job. I just had one big question around how part of it should work. I had a design sketch, but with some big unknowns. I asked fable to figure it out via reasoning and prototyping, and it did - it even, under its own initiative, wrote a fuzzer for its prototype which explored and verified that its reasoning was correct. It absolutely nailed it. And it found, and fixed, a couple bugs that I'd missed.

I'm sure its weaknesses will become apparent in time. But, wow this thing is a beast. Its the first time I'm reading the work of an LLM without spotting obvious weaknesses in its reasoning and code. I'm really impressed.

Re: Claude Fable 5

#735

Post by slaymaker1907 »

shruubi wrote: post #700

I have a theory, this is obviously based on speculation based on how Anthropic is treating Mythos and the whole media noise around it's dangers and who gets access to it. My theory is that Anthropic are banking on being the top model when the race to IPO finally reaches the finish line, and to do that they need to have the top model but not let any competitors see it or derive from it to have a comparable model in th…

That's definitely the case as model distillation is one of the explicit safety carveouts they mention. Though TBF, model distillation is also a big concern for general safety as distillation could allow you to have the model without the other guardrails. It's sort of a master key to the model.

Re: Claude Fable 5

#736

Post by Madmallard »

Dig1t wrote: post #588

>To release the model both safely and quickly, we’ve tuned these safeguards conservatively—they’ll sometimes catch harmless requests Why is everyone so okay with these companies intentionally gimping their AI and choosing who is allowed to know certain types of information in the name of safety? Can you imagine if Microsoft shipped a feature in their OS that watched what you did and shut down the computer if it detec…

I mean it's all political in the first place. That's unavoidable. What are we going to do about it?

Re: Claude Fable 5

#738

Post by zitoshi »

I'm in the midst of learning loop design.

For those more advanced and have used fable, does fable make learning this less or more necessary?

As in, can I now reliably give higher order problems like ... "we are missing a feature in this app to make it complete, what is it?"

Or should i still be quite specific with defining success in a clean metric based way.

Re: Claude Fable 5

#739

Post by yoyohello13 »

steve_adams_86 wrote: post #696

I'm using it to review recent work and it's doing a genuinely excellent job. This is a clear step up. Fewer decisions I have to guide it away from, faster conclusions on planning, more willing to go out of the way to make the correct decisions possible... This is really interesting. It feels like going from Sonnet to Opus, but, of course as a step up from Opus. This feels more like working with a competent peer than…

Same. I used it today to review my code and it came up with some genuinely good comments and suggestions and found a bug I didn’t think about. Quite a step up from opus. Although one code review took up 50% of my usage.

Re: Claude Fable 5

#740

Post by sigbottle »

simonw wrote: post #534

I've spent enough time with this now in Claude Code (and Claude.ai and Claude Code for web) to have an opinion on Fable 5: it's a beast. I'm throwing some VERY difficult problems at at - things I've been dragging my heels on for months - and it's crunching through them very happily. One that I'm willing to share (albeit from just a week ago) - I built a Python library last week that bundles MicroPython compiled to WA…

Does anyone know what the architecture of Fable is? Is it harnesses? Did they solve persistent learning? What did they do?
Post reply on HN