Skip to content
A

Pydantic AI

Runtime & OrchestrationOpen source · MITSelf-hosted

Type-safe Python agent framework that validates LLM output with Pydantic.

Pydantic AI is an MIT-licensed Python agent framework from the Pydantic team that validates model output against Pydantic models and can automatically retry when validation fails, bringing the library's type-safety guarantees to agent structured output.

Key facts

CategoryRuntime & Orchestration
LicenseMIT
LanguagePython
Self-hostedYes
Last verified2026-10-04
Repositoryhttps://github.com/pydantic/pydantic-ai

Markdown version (for LLMs)

Tools in this category