What is the TypeSafe Jev Playground?

A BYOK experimentation workbench for practicing Jev concepts: State, atomic questions, Noul, Score and Choice. It runs either in LLM Practice Mode with your existing provider keys or in Native Jev Mode if you hold authorized credentials.

Is this the real Jev model?

In practice mode, no: the README states clearly it does not run TypeSafe's proprietary weights or hosted infrastructure, and makes no equivalence claim. Native Jev Mode does execute against official Jev infrastructure, but requires authorized credentials.

Which providers can I use in practice mode?

Any of the commodity LLM keys the README lists: Groq, OpenAI, Anthropic, Google Gemini, or a local Ollama instance. The mode is designed for developers without official Jev access who still want to learn the concepts.

What does it cost to use?

In practice mode, costs go directly to your own LLM provider, or nothing if you use a free Groq key. In native mode, usage is billed to your official Jev organization account.

Do I need a Jev license to try it?

No for practice mode: any existing LLM API key works. Native Jev Mode requires authorized native Jev API credentials, which the README presents as the path for authoritative production evaluation with official guarantees.

What can I experiment with in the playground?

The README lists State, atomic questions and the Noul, Score and Choice question types with structured decisions. Its stated purpose is learning JEV concepts, drafting questions and criteria rubrics, and prototyping workflows.

Who built the playground and what stack does it use?

It is saimahmedqazi's project, built with TypeScript 5.7, React 19, Vite, Tailwind CSS and Express, hosted on Vercel at typesafe-jev-playground.vercel.app. The README badges MIT and a 102-test Vitest suite; both are self-reported.

Jev directory

CommunityJev Demos & Experiments

TypeSafe Jev Playground

A BYOK workbench for practicing Jev concepts without official access: emulates State, Noul, Score and Choice via Groq, OpenAI, Anthropic, Gemini or local Ollama, or runs against native Jev with authorized credentials.

See highlights Open github.com

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

saimahmedqazi/typesafe-jev-playground

User repository on GitHub

View on GitHub

A BYOK playground for practicing TypeSafe's Jev — the first System One model — with your own LLM or native Jev API access. Experiment with Noul, Score, Choice, custom state, and structured decisions.

Forks: 0
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

Tagscommunity catalog-import logicrw byok playground llm

Works forJev for Classification Jev for Scoring Jev for Evaluation

Used inJev for Software & DevTools

Languages: TypeScript
Platforms: Vercel AI Gateway Browser
Capabilities: Open sourceRequires API keyRuns locally

Highlights

Explicitly not the real Jev runtime: LLM Practice Mode emulates JEV workflows with commodity models so you can learn without official access.
Native Jev Mode executes against official TypeSafe infrastructure for developers with authorized credentials.
Covers State, atomic questions, Noul, Score and Choice; built with TypeScript 5.7, React 19, Vite 6.4, Tailwind and Express.
README claims Zero-Fallback BYOK and a 102-test Vitest suite; MIT per the README badge; hosted demo on Vercel.
Practice mode works with Groq, OpenAI, Anthropic, Gemini or local Ollama keys you already own.

Watch out

Grounded in the truncated README and GitHub metadata reviewed on 2026-10-15; the README was cut off inside the setup section, its badges, test count and Zero-Fallback BYOK claims are self-reported, the GitHub metadata lists no licence though the README badges MIT, and practice mode is an emulation with no equivalence claim.

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.

Per-Tick Decisions in Games

Sep 28, 2026

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.

Jev Demos & Experiments#gaming#agents#control

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai