index / themsquared/jev-benchmark
jev-benchmark
Reproducible benchmark for TypeSafe AI's Jev on agent tool-call risk classification: accuracy, latency, and whether the confidence score is worth routing on.
Research and EvalsVerificationListedPython
Repository
Stars0
LanguagePython
LicenseApache-2.0
Created2026-09-17
Last pushed2026-09-17
Officialno
Judgment
Genuine Jev project0.97
gate 0.5
Uses Jev at runtime0.86
Is a meta list0.03
Is a reimplementation0.06
Quality scales
Substance1.59 / 3
gate 0.5
Docs2.63 / 3
Novelty2.91 / 3
Composite score0.75
Category distribution
Research and Evals1.00
Learning0.00
Games and Simulation0.00
Integrations0.00
Other0.00
Applications0.00
Agent and Dev Tooling0.00
SDKs and Clients0.00
Assigned categoryResearch and Evals
Category probability1.00
Category confidence1.00
PatternVerification
Pattern confidence0.40
Curation
StatusListed
Reasongate passed
README pickno
Overridenone
Sources
- search:jev typesafe
- search:"typesafe.ai"
- search:jev in:name,description,readme created:2026-09-17
Provenance
Modeljev-1.13.0
Question setv2
Judged at2026-09-18 14:05 UTC