index / denser-org/rerank-bench-jev
rerank-bench-jev
Production reranker benchmark: TypeSafe Jev vs Qwen3-Reranker-0.6B on BEIR SciFact and NFCorpus — nDCG@10, cost/query and p50/p95 latency. Quality is indistinguishable; Jev is ~2x faster, Qwen ~4x cheaper. From the team at denser.ai
Research and EvalsRankingListedPython
Repository
Stars0
LanguagePython
LicenseApache-2.0
Created2026-09-22
Last pushed2026-09-22
Officialno
Judgment
Genuine Jev project0.95
gate 0.5
Uses Jev at runtime0.82
Is a meta list0.03
Is a reimplementation0.05
Quality scales
Substance1.98 / 3
gate 0.5
Docs2.43 / 3
Novelty2.83 / 3
Composite score0.78
Category distribution
Research and Evals1.00
Applications0.00
Learning0.00
Other0.00
SDKs and Clients0.00
Games and Simulation0.00
Agent and Dev Tooling0.00
Integrations0.00
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:2026-09-22
Provenance
Modeljev-1.13.0
Question setv2
Judged at2026-09-22 11:45 UTC