BerriAI/litellm · error · ValueError

api_base is required for Pydantic AI agents

Error message

api_base is required for Pydantic AI agents

What it means

PydanticAIProviderConfig.handle_streaming raises ValueError when api_base is falsy. Streaming for Pydantic AI is simulated (fake streaming) from a non-streaming HTTP call, which still requires the agent's base URL.

Source

Thrown at litellm/a2a_protocol/providers/pydantic_ai_agents/config.py:47

            raise ValueError("api_base is required for PydanticAIProviderConfig")
        return await PydanticAIHandler.handle_non_streaming(
            request_id=request_id,
            params=params,
            api_base=api_base,
            timeout=kwargs.get("timeout", 60.0),
            agent_extra_headers=kwargs.get("agent_extra_headers"),
        )

    async def handle_streaming(
        self,
        request_id: str,
        params: dict[str, Any],
        api_base: str | None = None,
        **kwargs,
    ) -> AsyncIterator[dict[str, Any]]:
        """Handle streaming request with fake streaming."""
        if not api_base:
            raise ValueError("api_base is required for Pydantic AI agents")
        async for chunk in PydanticAIHandler.handle_streaming(
            request_id=request_id,
            params=params,
            api_base=api_base,
            timeout=kwargs.get("timeout", 60.0),
            chunk_size=kwargs.get("chunk_size", 50),
            delay_ms=kwargs.get("delay_ms", 10),
            agent_extra_headers=kwargs.get("agent_extra_headers"),
        ):
            yield chunk

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Provide a non-empty api_base in the call/deployment config
2. Verify the api_base env/config variable resolves at request time (not an unset env default of '')

Example fix

# before
async for c in config.handle_streaming(request_id=rid, params=params, api_base=os.getenv("AGENT_URL")): ...
# after
base = os.environ["AGENT_URL"]  # fail fast if unset
async for c in config.handle_streaming(request_id=rid, params=params, api_base=base): ...

if not api_base:
    raise ValueError("pydantic-ai streaming requires api_base")

When it happens

Trigger: Calling handle_streaming on the pydantic_ai_agents config with api_base=None or empty string.

Common situations: Streaming requests through a gateway where the deployment lacks api_base; passing "" from an env var that was never set.

Related errors

• api_base is required for PydanticAIProviderConfig
• api_base is required for Pydantic AI agents
• Either a2a_client or api_base is required for standard A2A f
• Task {task_id} ended with state: {state}
• Task {task_id} did not complete within {max_attempts * poll_

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/87d64d942d36f27c. Report an issue: GitHub.