# Awesome Jev

> An evidence-graded, continuously rebuilt index of the Jev / TypeSafe
> System One ecosystem: SDKs, MCP tooling, agent guardrails, evaluations,
> open models, applications and writing. Every entry carries an evidence
> grade saying how much was actually verified. Rebuilt every two hours in
> twenty languages.

Last generated: 2026-09-23T15:10:39+08:00
Entries: 704

## What Jev is

Jev is TypeSafe AI's first System One model. It takes a state plus typed
questions whose answer space is fixed in advance and returns constrained
answers with probability distributions, rather than generating text. The
primitives are Choice, Score and Noul. Endpoint: POST
https://api.typesafe.ai/v1/systemone, model jev-latest.

## Evidence grades

- official: published by TypeSafe AI itself
- observed: found used in real code, or its own text cites the API
- inferred: declared in a repository name or an explicit topic tag
- unverified: matched on vocabulary alone

The grade describes how much was verified, not how good a project is.

## Sections

- agent-tooling: 224 entries
- community-sdk: 117 entries
- media-discussions: 92 entries
- evaluation: 70 entries
- other: 68 entries
- routing-guardrails: 54 entries
- apps-demos: 41 entries
- research-models: 31 entries
- official-sdk: 7 entries

## Pages

- [Index](https://wh000wh000.github.io/awesome-jev-live/): searchable, filterable list of every entry
- [Repository](https://github.com/wh000wh000/awesome-jev-live): the generated README, contributing guide and change ledger
- [English README](https://github.com/wh000wh000/awesome-jev-live/blob/main/README.md)
- [Chinese README](https://github.com/wh000wh000/awesome-jev-live/blob/main/docs/README.zh-CN.md)
- [Data](https://github.com/wh000wh000/awesome-jev-live/blob/main/data/entries.json): the underlying entries as JSON

## Disclosure

Independent community project, not affiliated with or endorsed by TypeSafe
AI. Product behaviour, pricing and model aliases change without notice.