KNotJev : Kev

Benchmarks

Same public held-out suite used around OpenJev: s1lv3rj1nx/openjev-heldout. Model: qwen3.5:9b via Ollama · 100% GPU · n=600 per task.

Why NotJev : Kev looks strong here

  • Banking77 at 83% — at or above published Jev held-out (~82%) and JevBench full-test 80.3%, on your own open stack.
  • High-cardinality choice works — CLINC OOS 86% across 151 intents with cascade / constraints.
  • All three primitives — choice, ordinal score (SST-5 within-1), and noul toxicity gates.
  • Reproducible & free to re-run — no proprietary weights required; point Kev at any model you control.
TaskPrimitiveKevChanceReference
Banking77choice (77)83%1.3%Jev ~82% held-out · JevBench 80.3%
CLINC OOSchoice (151)86%0.7%
AG Newschoice (4)86.5%25%
SST-5score (5)89.5% within-120%ordinal ±1
Civil Commentsnoul81%50%

OpenJev private 10k mix (cite only): Jev 85.4% · OpenJev 84.0% — questions not fully public, so we do not invent a score against it. SST-5 uses ordinal within-1; other tasks are exact-match.

Reproduce

KEV_BACKEND=ollama
KEV_OLLAMA_MODEL=qwen3.5:9b
pnpm --filter @kev-ai/server start
pnpm bench:heldout -- --mode api --base-url http://127.0.0.1:3000 \
  --tasks banking77,clinc_oos,ag_news,sst5,civil_comments_toxicity