Google Provider

class llm_async.providers.google.GoogleProvider(api_key: str = '', base_url: str = '', vertex_config: dict[str, Any] | None = None, retry_config: RetryConfig | None = None, client_kwargs: dict | None = None, http2: bool = False)[source]

Bases: BaseProvider

Google Provider supporting both Google AI API and Vertex AI.

Args:

api_key: API key for Google AI API (ignored if vertex_config provided) base_url: Base URL for API (auto-detected if vertex_config provided) vertex_config: Configuration for Vertex AI with keys:

  • project_id: Google Cloud project ID

  • location_id: Location/region (e.g., “us-central1”, “global”)

  • api_endpoint: Optional custom API endpoint

  • goth_token: Optional Bearer token for Vertex AI auth

async execute_tool(tool_call: ToolCall, tools_map: dict[str, Callable[[...], Any]]) → dict[str, Any][source]

Execute tools and return results in provider-specific format.