Python: [Bug]: model_id versus model_deployment_name is confusing in Azure AI Agents

[tonybaloney]

Description

In Foundry agents, the inconsistency between model_id and model_deployment_name is confusing. For example, if you don't provide the model_deployment_name argument to AzureAIClient it reports a ValueError when running an agent because model_id isn't set on the client object, even if it was set on the call to AzureAIClient.create_agent. model_id is accepted as a kwarg so you don't get a linter error.

AzureAIClient.create_agent takes model_id as a parameter instead of model_deployment_name.

Code Sample

Error Messages / Stack Traces

Package Versions

main

Python Version

No response

Additional Context

No response

[eavanvalkenburg]

Hey, thanks for reaching out, we are looking at how we define clients, so we do have a opportunity to make changes now, however the reason we use model_deployment_name for the Azure based clients is to make it easy to understand what you need to look for in the portal, since they all refer to that name. I did notice where the confusion comes from, so i will have a look at how to fix that.