Skip to content
Docs

Pydantic AI with AI Gateway

Build Python agents with Pydantic AI and connect them to AI Gateway for access to its model catalog and provider routing.

  1. First, create a new directory for your project and initialize it:

    terminal
    mkdir pydantic-ai-gateway
    cd pydantic-ai-gateway
  2. Install the required Pydantic AI packages along with the python-dotenv package:

    terminal
    pip install pydantic-ai python-dotenv
  3. Create a .env file with your Edgenivo AI Gateway API key:

    .env
    VERCEL_AI_GATEWAY_API_KEY=your-api-key-here

    If you're using the AI Gateway from within a Edgenivo deployment, you can also use the VERCEL_OIDC_TOKEN environment variable which will be automatically provided.

  4. Create a new file called main.py with the following code:

    main.py
    from dotenv import load_dotenv
    from pydantic import BaseModel
    from pydantic_ai import Agent
    from pydantic_ai.models.openai import OpenAIModel
    from pydantic_ai.providers.vercel import VercelProvider
     
    load_dotenv()
     
    class CityInfo(BaseModel):
        city: str
        country: str
        population: int
        famous_for: str
     
    agent = Agent(
        OpenAIModel('anthropic/claude-opus-5', provider=VercelProvider()),
        output_type=CityInfo,
        system_prompt='Provide accurate city information.'
    )
     
    if __name__ == '__main__':
        cities = ["Tokyo", "Paris", "New York"]
     
        for city in cities:
            result = agent.run_sync(f'Tell me about {city}')
            info = result.output
     
            print(f"City: {info.city}")
            print(f"Country: {info.country}")
            print(f"Population: {info.population:,}")
            print(f"Famous for: {info.famous_for}")
            print("-" * 5)

    The following code:

    • Defines a CityInfo Pydantic model for structured output
    • Uses the VercelProvider to route requests through the AI Gateway
    • Handles the response data using Pydantic's type validation
  5. Run your application using Python:

    terminal
    python main.py

    You should see structured city information for Tokyo, Paris, and New York displayed in your console.

Last updated September 8, 2026

Was this helpful?

supported.