# Jev Buddy

> The friendly developer guide to Jev, the AI decision model: choice, noul and score explained with code, business examples, a playground and a cost calculator.

Jev Buddy is a free, multipage developer guide to Jev, TypeSafe AI's System One decision model. You send it state (text or JSON) and typed questions (choice, noul and score), and it returns typed answers with calibrated probabilities instead of text.

The site explains everything a developer needs, with code in Python, TypeScript, curl and TypeSafe's SDK, shows twelve business use cases with routing code and costs, and has a playground, saved recipes and calculators. An Open source page shows how to install and use open alternatives such as Laya, Kev, jevos and Ollaya, so decisions run on your own machine and your data stays with you. It is an independent guide, not affiliated with TypeSafe or OpenRouter.

Who it's for: Developers deciding with AI: routing, scoring, classifying and gating

## What it does

- A 16-chapter guide with code in Python, TypeScript, curl and the SDK
- The three primitives explained: choice, noul and score
- Twelve business use cases with routing code and costs
- A playground and saved recipes
- Cost calculators
- Open-source, self-hosted alternatives to Jev (Laya, Kev, jevos, Ollaya) with install and usage steps

## Questions

### Is Jev Buddy official?

No. It is an independent guide, not affiliated with TypeSafe or OpenRouter.

### What is Jev?

A decision model: it takes state and typed questions and returns typed answers with calibrated probabilities, rather than generating text.

### Is the guide free?

Yes. Everything is readable without an account.

### Is there an open-source alternative to Jev?

Yes. Laya, Kev, jevos and Ollaya answer the same choice, yes/no and score questions on your own machine, and several accept the same request as TypeSafe's API. The Open source page has install steps for each.

## Links

- [Open Jev Buddy](https://jev-buddy.anvil.app/)
- [About Jev Buddy](https://jev-buddy.anvil.app/about)
- Made by [Arkvital](https://arkvital.com), Montreal, Canada (arkvital@outlook.com)

Keywords: Jev decision model, TypeSafe AI, AI classification API, LLM alternative for decisions, typed AI answers, calibrated probabilities API, AI routing, support ticket classification, AI cost calculator, open-source Jev alternative, Laya decision model, self-hosted decision model