Kev: open-source 0.8B/4B/9B judge models on Qwen3.5, the 9B fits a 32GB Mac
khiladi1729 · reddit · 2026-09-22
Kev is a small family of Jev-architecture decision models built on Qwen3.5, available in 0.8B, 4B and 9B sizes under Apache-2.0, with training code and frozen eval suites included.
- The 9B model runs locally on a 32GB Mac in bf16
- It's API-compatible with TypeSafe's System One, so their Python SDK points at your local server unchanged
- The bundled web playground demonstrates how option order shifts judge answers — the bias every LLM-as-judge builder runs into
Repo: https://github.com/jaredpalmer/kev
More from coding & agent
- Loom founder launches Roma 2.0, a shared to-do list that orchestrates AI agents — thetripathi58 · 2026-11-12
- Sosumi Eval: Apple Docs Access Lifts Coding Agent Pass Rates from ~33% to 94-98% — JordanMorgan10 · 2026-11-12
- CypherGuard fixes structurally wrong LLM Cypher queries, hitting 88.5% on CypherBench — JeremyCMorgan · 2026-11-12
- Building an End-to-End Hedge Fund in Python with Claude Code: A 5-Step Workflow — mdancho84 · 2026-11-12
- Jeff Dean: comfortable with engineers spending up to $10M/year on AI tools — deedydas · 2026-11-12
- Developer builds VladOS, a personal life dashboard, by just describing it to an AI agent — haltakov · 2026-11-12