Fable vs Opus 5: My Rule

People keep asking me whether Opus 5 or Fable 5 is the better Claude model.

That's the wrong question, and answering it wrong costs real money every month.

Every model launch comes with benchmark charts and confident hot takes, so business owners pick whichever sounds smarter. Then they pay double for a model that spends its day answering emails.

I treat the two models like hires. You wouldn't ask your CEO to clear the inbox, and you wouldn't ask your intern to decide the future of the company. So I give Fable the big decisions and Opus the daily work.

Fable 5 is built for the long run. Give it a task that takes hours or even days and it holds the thread the whole way through. It remembers what it decided three steps back, and it can manage other AI agents working underneath it without losing the plot. That's the model for your hardest problem and your highest stakes call.

Opus 5 is built for right now. It's faster, it costs half as much, and for most of what a business needs every day, drafting an email, writing a proposal, pulling a report together, it holds up just as well.

The hiring test does the deciding for you. A quarterly pricing review is CEO work, so it goes to Fable. The Tuesday follow-up email is intern work, so it goes to Opus.

If the job is long or the stakes are high, open Fable. If it's daily work you need in the next ten minutes, open Opus. Pick the model for the job in front of you, not the one with the smarter-sounding name.