UrlContextTool: '400 Tool use with function calling is unsupported' if `output_type` is specified

[nvzoll]

Initial Checks

• I confirm that I'm using the latest version of Pydantic AI
• I confirm that I searched for my issue in https://github.com/pydantic/pydantic-ai/issues before opening this issue

Description

Trying to use UrlContextTool gives me an error

google.genai.errors.ClientError: 400 INVALID_ARGUMENT. {'error': {'code': 400, 'message': 'Tool use with function calling is unsupported', 'status': 'INVALID_ARGUMENT'}}

if I have output_type for the agent specified. When I remove output_type, it runs as expected.

Example Code

from pydantic_ai import Agent, UrlContextTool
from pydantic_ai.models.google import GoogleModel
from pydantic_ai.providers.google import GoogleProvider

GEMINI_API_KEY = '...'

agent = Agent(
    model=GoogleModel(
        "gemini-2.5-flash",
        provider=GoogleProvider(api_key=GEMINI_API_KEY),
    ),
    output_type=MyPydanticModel, # If we comment out this line, it works as expected
    system_prompt='system prompt here',
    builtin_tools=[UrlContextTool()],
)

res = await agent.run('user prompt here')

Python, Pydantic AI & LLM client version

Python 3.13.2
Pydantic AI 0.8.1
LLM gemini-2.5-flash, provider=GoogleProvider

[nvzoll]

(I understand that it's probably not pydantic.ai's fault, but it would be nice to at least specify this outcome in the docs.)

[DouweM]

@nvzoll That's a very unfortunately limitation. I've created a PR to add a clearer error message and a docs mention.