JevMade hello@JevMade.com
← Back to experiments

Benchmarks & research

jevals

A Python library for checking an agent's work with several Jev questions in one call.

Source screenshot of jevals
SOURCE SCREENSHOT · source ↗ · captured 2026-09-22Full screenshot ↗

Maker-reported (not independently measured by JevMade): One request for eight evals returned 1,388 tokens, $0.00006 and 0.33 s in the README's quickstart trace (author-reported) · p50 244 ms and p95 371 ms per request measured through Vercel's gateway (author-reported)

Primitives
choice, noul
Platform
Python
Added
Project created
GitHub stars
45 · snapshot 2026-09-22

Source checked 2026-09-22 — opened the primary source directly.