index / laguagu/jev-rerank-bench
jev-rerank-bench
Does a typed-judgment model rerank better than a cross-encoder? Jev, Voyage, GPT-4.1-mini, GPT-5.6-luna and Laya over identical shortlists on two Finnish legal corpora.
Research and EvalsRankingListedTypeScript
Repository
Stars0
LanguageTypeScript
LicenseMIT
Created2026-09-20
Last pushed2026-09-20
Officialno
Judgment
Genuine Jev project0.97
gate 0.5
Uses Jev at runtime0.92
Is a meta list0.03
Is a reimplementation0.05
Quality scales
Substance1.89 / 3
gate 0.5
Docs2.53 / 3
Novelty2.88 / 3
Composite score0.78
Category distribution
Research and Evals1.00
Applications0.00
Games and Simulation0.00
Integrations0.00
SDKs and Clients0.00
Learning0.00
Agent and Dev Tooling0.00
Other0.00
Assigned categoryResearch and Evals
Category probability1.00
Category confidence1.00
PatternRanking
Pattern confidence1.00
Curation
StatusListed
Reasongate passed
README pickno
Overridenone
Sources
- search:jev in:name,description,readme created:2026-09-20
- code:"@typesafe-ai/sdk"
Provenance
Modeljev-1.13.0
Question setv2
Judged at2026-09-21 12:35 UTC