Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

jev-java (gudcks0305)

Multi-provider Java SDK for Jev via TypeSafe, OpenRouter, Vercel or Cloudflare, plus OpenAI Decisions and Spring Boot support.

LibrarybetaMIT
View on GitHub ↗

Overview

Jev Java (gudcks0305) is an unofficial Java 17+ SDK for turning application state into typed Jev judgements through TypeSafe, OpenRouter, Vercel AI Gateway or Cloudflare. You define Choice, Noul and Score questions in Java, submit them together and receive typed results, including mapping outputs onto Java records. The plain SDK uses the JDK HttpClient and Jackson; Spring is optional.

It is split into modules: jev-typesafe, jev-cloudflare, jev-openrouter, jev-vercel, jev-openai (version 0.3.0 adds OpenAI Decisions with text and image inputs, typed choices, scoring, per-question refusals and safety identifiers), a Spring Boot starter and an optional WebFlux transport. It is not affiliated with TypeSafe, Vercel or OpenAI.

What the decision model decides

TypeQuestionAnswers
OtherAny Choice, Noul or Score question mapped onto a Java record—

How it works

  1. Add the provider module for your route
  2. Declare questions or a Java record for output
  3. Submit together
  4. Receive typed results

Strengths

  • Many providers in one SDK
  • Record mapping
  • Spring Boot and WebFlux support

Caveats

  • Unofficial
  • OpenRouter and Vercel protocols are alpha or experimental

Getting started

Add `io.github.gudcks0305:jev-typesafe:0.3.0` from Maven Central.

At a glance

Type
Library
Status
beta
Language
Java
Runtime
Java 17+, Spring Boot optional
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

Which providers are supported?

TypeSafe, OpenRouter, Vercel AI Gateway, Cloudflare and OpenAI Decisions.

Does it work with Spring?

Yes, via a Spring Boot starter and optional WebFlux transport.

Can outputs map to records?

Yes, record mapping is built into jev-core.

Related projects

  • Ax

    Ax, a DSPy-style multi-language LLM framework, listed with a native TypeSafe provider for Boolean, Choice, Score and raw Jev questions.

    PluginproductionTypeScript
  • Effect

    Effect TypeScript library whose @effect/ai-typesafe package maps classify, probability and rating operations to Jev questions.

    ProjectproductionTypeScript
  • instructor-php

    PHP monorepo whose Polyglot module adds a TypeSafe driver for typed Noul, Choice and Score decisions alongside LLM inference.

    ProjectproductionPHP
  • req\_llm

    Elixir library for unified LLM calls across providers, with an evaluate interface that calls Jev via TypeSafe or OpenRouter.

    ProjectproductionElixir
  • synkora-ai

    Self-hosted multi-tenant AI agent platform listed as including optional TypeSafe client tools for classification, scoring and yes/no.

    LibraryproductionPython
  • TanStack AI

    Provider-agnostic TypeScript AI SDK; per list sources, an @tanstack/ai-typesafe adapter exposes Jev decisions through decide().

    ServiceproductionTypeScript