Pydantic AI
A Python framework for building AI agents, developed by the team behind the Pydantic validation library, featuring Code Mode for LLM-generated script execution.
Pydantic AI is an agent framework that extends the popular Pydantic data validation library into the AI agent space. It offers conventional tool-calling patterns alongside an experimental Code Mode, where LLMs generate Python scripts that chain multiple operations together. The framework includes Monty, a minimal Rust-based Python interpreter designed for safe execution of LLM-generated code with microsecond startup times.
Also known as
PydanticAI, Pydantic Agent, Monty interpreter