> ## Documentation Index
> Fetch the complete documentation index at: https://phidatainc.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# PydanticAI Instantiation Performance Evaluation

> Benchmark 1000 PydanticAI agent constructions, including an inline @agent.tool_plain weather tool, via Agno's PerformanceEval.

Demonstrates agent instantiation benchmarking with PydanticAI.

```python pydantic_ai_instantiation.py theme={null}
"""
PydanticAI Instantiation Performance Evaluation
===============================================

Demonstrates agent instantiation benchmarking with PydanticAI.
"""

from typing import Literal

from agno.eval.performance import PerformanceEval
from pydantic_ai import Agent


# ---------------------------------------------------------------------------
# Create Benchmark Function
# ---------------------------------------------------------------------------
def instantiate_agent():
    agent = Agent("openai:gpt-4o", system_prompt="Be concise, reply with one sentence.")

    # Tool definition remains scoped to agent construction by design.
    @agent.tool_plain
    def get_weather(city: Literal["nyc", "sf"]):
        """Use this to get weather information."""
        if city == "nyc":
            return "It might be cloudy in nyc"
        elif city == "sf":
            return "It's always sunny in sf"
        else:
            raise AssertionError("Unknown city")

    return agent


# ---------------------------------------------------------------------------
# Create Evaluation
# ---------------------------------------------------------------------------
pydantic_instantiation = PerformanceEval(func=instantiate_agent, num_iterations=1000)

# ---------------------------------------------------------------------------
# Run Evaluation
# ---------------------------------------------------------------------------
if __name__ == "__main__":
    pydantic_instantiation.run(print_results=True, print_summary=True)
```

## Run the Example

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U agno memory-profiler pydantic-ai
    ```
  </Step>

  <Step title="Export your API keys">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
      $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Run the example">
    Save the code above as `pydantic_ai_instantiation.py`, then run:

    ```bash theme={null}
    python pydantic_ai_instantiation.py
    ```
  </Step>
</Steps>

Full source: [cookbook/09\_evals/performance/comparison/pydantic\_ai\_instantiation.py](https://github.com/agno-agi/agno/blob/main/cookbook/09_evals/performance/comparison/pydantic_ai_instantiation.py)