Kev refactored onto Qwen3.5: open-source decision models now at 0.8B, 4B and 9B
alexcovo_eth · x · 2026-09-22
Jared Palmer announced that Kev, a family of small open-source Jev-like decision models, has been refactored on top of Qwen3.5, with new checkpoints at 0.8B, 4B and 9B plus a fine-tuning script that runs on Modal.
Kev uses the same LoRA + small pointer head technique as before. The prior Kev-8B scored 79.6% on out-of-domain data vs. Jev's 85.7%. The API is a drop-in TypeSafe System One replacement via a single baseurl change; Kev-4B serves on a 32GB Mac in bf16 (300ms for five questions, 40ms on H100), repeated documents hit a KV cache for 2-2.5x speedup, and it's Apache 2.0 — Kev-4B trains in 40 minutes on one H100.
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