Anthropic's own product team already tells enterprise customers to run cheap models like Sonnet and Haiku by default and only escalate to the flagship model when tasks get too hard, which is model routing driven by the customer.
Pash explains that Anthropic already promotes a routing pattern where Sonnet/Haiku handle classification and customer-service tasks while Fable is pulled in only for harder problems, meaning enterprise-driven model routing already exists. ✦ AI generated
Pash · The Cognitive Revolution · 2026-07-02 · original ↗
starts at this moment · 28:00
They've been promoting to people if you're going to use Sonnet and Haiku use Sonnet and Haiku in your kind of classification systems for your customer service but when things get too difficult pull in a fable. And so this is basically model routing but model routing driven by the enterprise customer themselves.
verbatim transcript · starts at 28:00
28:00an adviser for uh various tasks and in fact on the API that is actually what the product team at anthropic has been promoting. They've been promoting to people if you're going to use Sonnet and Haiku use Sonnet and Haiku in your kind of classification systems for your customer service but when things get too difficult pull in a fable. And so this is basically model routing but model
28:25routing driven by the you know enterprise customer themselves teaching them how to model route effectively using a smaller model. But you can also model it out using a uh using a better model and that is part of this kind of uh split tasks into to-do lists and then send out independent agents. Uh some people have been on fable right now on in in claude code if you set it to ultra
28:53code right now you actually get this experience. uh but they don't Fable doesn't use like uh sub agents or you don't know whether they're using sub aents which are of lower models. They won't tell you but you can explicitly ask Fable to use sonnet sub aents and it will use those sonnet sub aents and check their work and so you can start like streaming a lot of this work
29:15actually to other agents even using the existing model and then cut down. I hit my fable limit yesterday, the the five hour limit. I hit my fable limit and I went online and um you know I saw one of the other uh you know uh software engineers, influencers online and he said he had done 16 pull requests since Fable was out hadn't hit his limit at all but he had a structured strategy of
29:45he'd already told Claude these are the models that you can use. you have a fable, you have this, you have uh you and you have GPD 5.5 which is available to you by command line as a MCP tool and he basically told it and he also told it this is the cost and this is the uh effectiveness and this is the taste and then he told it depending on what cost