Structure Result of List throws exception

[ebamberg]

Initial Checks

• I confirm that I'm using the latest version of Pydantic AI

Description

When my structured result is a single result everything works fine.
But when my system prompt returns a list of results - like "Ensure the output are 5 search queries matching the user input." the framework throws an exception:
│ /home/erik/miniconda3/envs/llm/lib/python3.12/site-packages/pydantic_ai/_agent_graph.py:81 in │
│ increment_retries │
│ │
│ 78 │ def increment_retries(self, max_result_retries: int) -> None: │
│ 79 │ │ self.retries += 1 │
│ 80 │ │ if self.retries > max_result_retries: │
│ ❱ 81 │ │ │ raise exceptions.UnexpectedModelBehavior( │
│ 82 │ │ │ │ f'Exceeded maximum retries ({max_result_retries}) for result validation' │
│ 83 │ │ │ ) │
│ 84 │

╰────────────────────────────────────────────────────────────────────────────────────────────────────╯
UnexpectedModelBehavior: Exceeded maximum retries (1) for result validation

It does not matter how I define the list for example:

result_type=list[Person]
or
@DataClass
Answer:
persons: List[Person]
result_type=Answer

Example Code

@dataclass
class Person :
    name: str
    age: int       


agent = Agent(  
            model=llm,
            result_type=list(Person),
            system_prompt='Extract name and age of all persons from the following text. \n\n###TEXT####\n',  
        )

        result=agent.run_sync("John Doe, 36 years old, is an carpenter. Jane Austin,46 years old, is an author.")

Python, Pydantic AI & LLM client version

ollama (=OpenAPI client)
python 3.12
pydantic 0.0.26