Benchmarks & research

jev-eval

Compare Jev with general language-model judges on one multilingual booking-inquiry routing task.

What it does

The benchmark contains 60 synthetic tourism-photo inquiries in four languages; its live demo intentionally uses a different calling pattern from the measurement protocol.

Repository ↗Pinned implementation ↗Pinned README ↗

Primitives: choice, score
Platform: TypeScript
Added: Sep 21, 2026
Project created: Sep 18, 2026
GitHub stars: 0 · snapshot 2026-09-21

Source checked 2026-09-21 — opened the primary source directly.