AbdelStark / jev-benchmarksView on GitHub
Probability-aware evaluation for typed decision models: calibration, selective risk, latency, and reproducible benchmarks.
☆26Sep 17, 2026Updated last month

Alternatives and similar repositories for jev-benchmarks

Users that are interested in jev-benchmarks are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?