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

Vercel AI SDK provider

Official Vercel AI SDK provider @ai-sdk/typesafe-ai for calling Jev as an evaluation model with experimental_evaluate.

Librarybeta
Open source ↗

This summary is based on how curated lists describe the project; its README was not available to review.

Overview

The Vercel AI SDK has a TypeSafe AI provider package, @ai-sdk/typesafe-ai. According to the list summaries, you use it with the SDK's experimental_evaluate function and create an evaluation model with typeSafeAi.evaluationModel('jev-latest'), or use the AI Gateway id typesafe-ai/jev instead.

This lets TypeScript developers already using the AI SDK add typed Jev decisions alongside their generative model calls. The evaluate API is marked experimental.

What the decision model decides

TypeQuestionAnswers
OtherTyped evaluation questions via experimental_evaluate(inferred)—

How it works

  1. Install @ai-sdk/typesafe-ai.
  2. Create typeSafeAi.evaluationModel('jev-latest') or use typesafe-ai/jev via the Gateway.
  3. Call experimental_evaluate with state and questions.
  4. Use the typed results.

Strengths

  • Native integration with the Vercel AI SDK.
  • Works directly or through AI Gateway.

Caveats

  • The evaluate API is experimental.
  • Known only from list summaries here.

Getting started

npm install @ai-sdk/typesafe-ai, then use typeSafeAi.evaluationModel('jev-latest') with experimental_evaluate.

At a glance

Type
Library
Status
beta
Language
TypeScript
Runtime
Vercel AI SDK
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

What is the package name?

@ai-sdk/typesafe-ai.

How do I create the model?

typeSafeAi.evaluationModel('jev-latest'), or the Gateway id typesafe-ai/jev.

Is the API stable?

It uses experimental_evaluate, which is experimental.

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