Python docs: Errors with configure Azure AI Agent Client observability

[pamelafox]

I tried to follow this:
https://learn.microsoft.com/en-us/agent-framework/tutorials/agents/enable-observability?pivots=programming-language-python#azure-ai-foundry-integration

agent_client = AzureAIAgentClient(
    credential=AzureCliCredential(),
    project_endpoint=f"https://{os.environ['AZURE_AI_PROJECT']}.foundry.azure.com"
)
agent_client.setup_azure_ai_observability(enable_sensitive_data=True, exporters=[ConsoleSpanExporter()])

However, I got an error:

agent_framework.exceptions.ServiceInitializationError: Azure AI model deployment name is required. Set via 'model_deployment_name' parameter or 'AZURE_AI_MODEL_DEPLOYMENT_NAME' environment variable.

I don't know why I'd need a model for observability to work?

Also, I was generally confused by this section: can I use this approach for any kind of agent, or is this only for hosted AI Foundry agents? That'd be great to clarify.