What it is
A curated list of resources for Jev, TypeSafe AI's System One model that returns typed, calibrated decisions instead of text. Maintained by RoboKrunch, who benchmark Chinese edge-AI hardware and run real-measurement Jev experiments.
How it uses Jev
Jev takes structured state and typed questions (choice, noul, score) and returns calibrated probabilities over predefined options in 70–500ms. The list catalogs projects where Jev makes decisions: browser agents pick operations and elements, guardrails score blast radius, and extensions use returned probabilities to filter content.
Primitives:choicescorenoul
Technique worth stealing
Use typed questions with predefined options to get calibrated probabilities instead of generated text.
Try it
Visit typesafe.ai for $5 starter credits; see Quickstart for API details.
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricunclear0.59
- Shows a System One patternno0.20
- Handles uncertaintyno0.04
- Measuredno0.09
- Runnableunclear0.47
- Worth recommendingno0.19
- Model replicano0.04
- Problem scopescore on a 0–2 scale1.03
- About Jevyes0.95
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 23 Sept 2026.