What is Awesome Jev Lab?

A curated map of Jev and TypeSafe System One resources combined with independent, reproducible benchmarks: an ecosystem guide, tests, raw artifacts and the scripts needed to reproduce the results.

Is it an official TypeSafe project?

No. The README states it is an independent community project, not affiliated with or endorsed by TypeSafe, even though it links to official documentation and repositories.

What are the Noul, Choice and Score primitives?

They are how Jev turns natural-language state into typed decisions: Noul for yes/no, Choice for selecting an option and Score for ordered ratings. The lab benchmarks these primitives with reproducible artifacts.

What evidence quality should I expect?

The repo reports 38 passing tests on Python 3.12, tested against jev-1.13.0, with the lab stdlib only. It ships a validation report and a 28-case audit whose evidence labels separate inspectable implementations from author-reported results.

What integrations does it track?

The Python and JavaScript SDKs, official agent skills, the System One adapter, Vercel AI Gateway, a LangChain integration guide, OpenRouter hosted access, and the Cline browser harness.

Where should I start reading?

The Start here table points to the TypeSafe Quick Start for understanding Jev in five minutes, official patterns for architecture, curated use cases, the Jev Lab validation report to check claims, and the run-the-lab instructions to reproduce.

What are the limits of this entry?

The digest's README excerpt is truncated partway through the curated use-case section, so the complete 28-case audit and the run-the-lab commands were not available to ground here; GitHub metadata also lists no license.

Jev directory

CommunityJev Benchmarks & Evaluations

Awesome Jev Lab

An independent research lab and ecosystem map for Jev: reproducible benchmarks of the Noul, Choice and Score primitives with raw artifacts and scripts, plus a curated 28-case audit of SDKs, integrations and real-world usage.

See highlights Open github.com

Stars: 1
Language: Python
Reading time: 2 min
Published by: Community
Author: llt22
Last verified: Oct 2026

llt22/jev-lab

User repository on GitHub

View on GitHub

Hands-on research lab for TypeSafe's Jev (System One model): reproducible benchmarks of Noul/Choice/Score primitives, confidence gating, fan-out latency, agent control — plus a living audit of the Jev ecosystem.

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity catalog-import logicrw benchmarks awesome-list

Works forJev for Evaluation Jev for Decision gating

Used inJev for Software & DevTools

Languages: Python
Platforms: CLI
Capabilities: Open source

Highlights

Combines an ecosystem guide with independent tests: 38 tests passing, Python 3.12, tested against jev-1.13.0, and the lab itself is stdlib only.
Typed decisions are the framing: Noul for yes/no, Choice for selecting an option and Score for ordered ratings, benchmarked to be reproduced with raw artifacts.
Ships a validation report and a 28-case audit that labels evidence, separating inspectable implementations from author-reported results.
The integrations table tracks the Python and JS SDKs, agent skills, the System One adapter, Vercel AI Gateway, LangChain, OpenRouter and Cline's harness.
States plainly that it is an independent community project, not affiliated with or endorsed by TypeSafe.

Watch out

Copy reviewed against the truncated README on 2026-10-15; the curated use-case section cuts off mid-heading, so the fuller 28-case audit was not reviewed here. GitHub metadata lists no license for the repository.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

The trust ledger: measuring confidence your domain has earned

Oct 16, 2026

ASSAY-001 calibrated Jev on CLINC150 but found it overconfident on Banking77 — one audit, two verdicts. How calibration research measures the model-and-domain pair, discounts stated confidence, and signs the evidence.

Jev Benchmarks & Evaluations#calibration#evaluation#open-models

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai