OPEN-SOURCE PROJECT

jev-benchmarks

A benchmark comparing Jev and GLiNER on text classification, probability calibration and selective automation.

Stars
12
Forks
1
License
Apache-2.0
Last commit
2026-09-17

What Jev does here

Runs the same labeled text tasks through both backends and records probabilities, latency and failures.

Helps examine task-specific accuracy and whether confidence scores support chosen thresholds.

The author reports a 300-example pilot. Hosted Jev and local GLiNER timings are not hardware-normalized; this site did not rerun it.