- Category
- Developer Tools
- Pricing
- Open Source
- Type
- TOOL
- GitHub
- 2.7k stars
- Latest release
- kev-family
- Date
About
Kev is a family of small decision models (0.8B, 4B, and 9B parameters) built on Qwen3.5, designed to be trained and run locally rather than through a hosted service. It answers yes/no, multiple-choice, and rating questions in a single request over shared input text, and exposes an API compatible with TypeSafe's System One SDK so existing clients can point at a self-hosted server.
Why it made the leaderboard
Kev gives engineers a small, self-trainable decision-model family (0.8B/4B/9B) that's API-compatible with a hosted service, so teams can swap in their own trained classifier instead of depending on a closed third-party API.
Tags
decision-modelsqwenllmopen-sourceclassificationself-hosted
Tech Stack
Python
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.