Claude Opus 5 • Hacker News Board

Re: Claude Opus 5

#31

Post by danielbln » Fri, Jul 24, 2026, 5:09 PM UTC

datakan wrote: post #6

> Claude Opus 5 is not more capable overall than our most capable general-access model, Claude Fable 5 Ok then so what's the point?

Cost.

Re: Claude Opus 5

#32

Post by the_lucifer » Fri, Jul 24, 2026, 5:09 PM UTC

Noticed none of the comparisons mention Kimi K3. Is there a comparison chart?

Re: Claude Opus 5

#33

Post by SketchySeaBeast » Fri, Jul 24, 2026, 5:10 PM UTC

hoppp wrote: post #26

atraac wrote: post #23

Great that there's a new model but they could fix their existing infra. We're considering dropping our Claude Team sub cause it's unusable recently. Constant bugs, dropped sessions, issues switching models, http errors. It's becoming ridiculous

Funny that a company selling an AI software developer can't use it to fix their infra. Fixing those issues still requires humans.

Let's be honest - they're also still hiring software devs. AI still requires skilled humans in the loop and that's not going away.

Re: Claude Opus 5

#34

Post by varispeed » Fri, Jul 24, 2026, 5:10 PM UTC

vidarh wrote: post #18

datakan wrote: post #6

> Claude Opus 5 is not more capable overall than our most capable general-access model, Claude Fable 5 Ok then so what's the point?

Fable is twice the price.

If Fable gets correct answer quicker, then you might pay less than doing back and forth with Opus, plus you lose more of your own time.

I see no reason for using less able models in my workflows. There is this saying, penny wise and pound foolish

Re: Claude Opus 5

#35

Post by pmg1991 » Fri, Jul 24, 2026, 5:10 PM UTC

Same cost as 4.8 but better that 4.8. Happy to get more efficient model. But is there any reason all companies are releasing models back to back after GLM 5.2.

Re: Claude Opus 5

#36

Post by SubiculumCode » Fri, Jul 24, 2026, 5:10 PM UTC

thewebguyd wrote: post #30

> Opus 5’s safeguards match those of Claude Fable 5’s, with one change: it now permits source-code vulnerability discovery at all access levels. This means that the model can support defensive cybersecurity work while still blocking vulnerability discovery in compiled binaries, which is more commonly used offensively Why can't they also allow Fable to do so also? Why is source-code vulnerability discovery limited to…

Because they are a private company and get to do what they want.

Re: Claude Opus 5

#37

Post by taf2 » Fri, Jul 24, 2026, 5:10 PM UTC

eager to see how it benchmarks on https://deepswe.datacurve.ai/

Re: Claude Opus 5

#38

Post by not_a9 » Fri, Jul 24, 2026, 5:11 PM UTC

> Opus 5’s safeguards match those of Claude Fable 5’s, with one change: it now permits source-code vulnerability discovery at all access levels. This means that the model can support defensive cybersecurity work while still blocking vulnerability discovery in compiled binaries, which is more commonly used offensively.

Okay so it’s worse than Opus 4.8 for my purposes I guess?

Re: Claude Opus 5

#39

Post by siwakotisaurav » Fri, Jul 24, 2026, 5:11 PM UTC

rb2e wrote: post #17

https://www.anthropic.com/news/claude-opus-5 - A blog post for those not wanting to go through a 190ish page pdf

Thanks for that, looks really good. I can see why they were constantly pushing back fable going out of the max sub with these benchmarks

Re: Claude Opus 5

#40

Post by cebert » Fri, Jul 24, 2026, 5:11 PM UTC

I am very confused about what the difference between Opus 5 and Fable 5 is now. What is the purpose of having two models that are so similar? The main differences I see are cost and marginal capability, according to the Anthropic-provided benchmarks.