Jevals: an LLM evals framework that uses Jev as the judge

Sai Yashwanth @yashwanthsai29Originally posted 2026.09.18

Open original on X

I built Jevals, an evals framework for LLMs and AI agents. Instead of an LLM judge, grading runs on Jev. So every pass/fail comes with a calibrated confidence you can set a bar on. Cheaper and Faster.

This post quotes another post: “ChatGPT co-inventor releases Jev, a model built for decisions rather than text”