Jev Model Routers: Send Each Request to the Right Model

Routing is a classification problem: look at a request and pick a tier. Using a frontier LLM to make that call can cost more than the savings. Jev returns the choice plus a confidence score, so a router can escalate to a stronger model only when it is unsure.

Measure on your own traffic before trusting reported savings. A router's value depends on how often it picks a tier that fails and has to retry.

3 projects

jev-router

★ 98

gargpratyush/jev-router

Per-turn model routing for Claude Code and Codex. Simple work goes to the fast tier and difficult work to the strong tier.

JavaScript

safer-with-jev

★ 3

andrelandgraf/safer-with-jev

Neon Function proxy for the Neon AI Gateway. Jev classifies each request and routes it to the right downstream model.

TypeScript

tiershift

★ 2

iamvatsalpatel/tiershift

Sends each LLM request to the cheapest model tier that can handle it and escalates on evidence.

TypeScript

Other categories

Get new Jev projects every week

New Jev releases, pricing changes, and the best new projects, once a week. No spam; unsubscribe anytime.