I'm shocked by how easy it is to run PydanticAI with a model running logically.

Here we have Ollama running llama3.2 locally, being used by @pydantic.dev AI extracting structured data from an input.

I take enormous pleasure in this: "uv run" is 6 characters same as "python", so you can switch between them and there's no jump.

One of uv's best features.

ai.pydantic.dev/examples/