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

ai-python

Vercel's AI SDK for Python, in public beta, listed for carrying Jev through its evaluation operation and AI Gateway examples.

Librarybeta
View on GitHub ↗

Overview

The AI SDK for Python is Vercel's toolkit for building LLM applications and agent loops in Python, currently in public beta. It wraps provider APIs as model config objects, routes provider IDs through AI Gateway by default, and supports direct OpenAI- and Anthropic-compatible providers via extras, plus tools and agents with streaming events.

The list summaries say the SDK carries Jev through its evaluation operation and Gateway examples. The README excerpt available here does not show that operation, so its exact API is not described. Models are thin config objects passed to ai.stream, agents run tools and stream typed events, provider IDs without a prefix route through AI Gateway, and direct OpenAI- or Anthropic-compatible providers load their official SDKs as optional extras.

What the decision model decides

TypeQuestionAnswers
OtherTyped evaluation questions passed through the SDK's evaluation operation(inferred)—

How it works

  1. Install the SDK
  2. Select Jev through AI Gateway
  3. Call the evaluation operation with state and questions
  4. Use the returned probabilities in Python code

Strengths

  • Official SDK with Gateway routing
  • Same toolkit for generation and evaluation

Caveats

  • Public beta
  • Jev operation not shown in the README excerpt

Getting started

uv add ai

At a glance

Type
Library
Status
beta
Language
Python
Runtime
library
Has code
yes

Use cases and integrations

How curated lists describe it

  • Python ecosystem: the official Vercel AI SDK for Python carries Jev through its evaluation operation and Gateway examples.

    yibie/awesome-jev

Frequently asked questions

How does it relate to Jev?

Per the listing, Jev is reachable through its evaluation operation and AI Gateway examples.

Is it stable?

It is in public beta.

How do I install it?

uv add ai, with extras for direct providers.

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