Routing with Jev: let Claude think, let Jev decide what deserves thinking
Jev returns only a choice, a score or a yes/no with a probability, at 70-500ms and $0.042 per million input tokens, and is used to route work to expensive models.
we've been using Claude f*cking wrong all the time your $200/mo model writes code, researches, plans... and then burns the same brain on YES or NO. Jev flips that. it's a model that cannot write a single sentence. you give it a situation and a question, it gives back one of three things: -> a choice from your list -> a score -> a yes / no each with a probability, so your code knows how sure it was. that's the whole product. -> 70 to 500 milliseconds per answer -> $0.042 per million input tokens -> output is free -> 10,000 decisions for around $0.42. that's not because Jev became smarter than Claude. it's because nobody asked it to be Claude. > let Claude think. > let Jev decide what deserves Claude. > let code do the actual work. the catch: text only, and still behind a waitlist. where to wire this into an agent today + the exact router setup in the article below
This post quotes another post: x.com/thegreatest_sv/status/2102802378378621355