Python: Bug: response_format not supported with AzureAIInferenceChatPromptExecutionSettings

[ManniArora]

Describe the bug
Response format is not support with AzureAIInferenceChatPromptExecutionSettings. Azure AI Inference chat completion API supports response_format but prompt execution settings do not support this.

For reference: https://github.com/Azure/azure-sdk-for-python/blob/372556c19c9ee3e3b5c1338c4be5eb287d902472/sdk/ai/azure-ai-inference/azure/ai/inference/aio/_patch.py#L142

To Reproduce
None of the below pass response_format to the underlying completions call -

response_format = JsonSchemaFormat(
            name="Recipe_JSON_Schema",
            schema=Recipe.model_json_schema(),
            description="",
            strict=True,
        )

AzureAIInferenceChatPromptExecutionSettings(
            service_id="service-id", response_format=response_format
        )

AzureAIInferenceChatPromptExecutionSettings(
            service_id="service-id", extra_parameters={"response_format": response_format}
        )

Expected behavior
Screenshots

Platform

• Language: Python
• Source: pip package version 1.26.1
• AI model: [e.g. OpenAI:GPT-4o-mini(2024-07-18)]
• IDE: [e.g. Visual Studio, VS Code]
• OS: [e.g. Windows, Mac]

Additional context
Add any other context about the problem here.

[moonbox3]

@ManniArora As of last month the Azure AI inference SDK didn't support response_format. Now, as you point out, they do. This is just a gap. We'll work on getting it handled.

[moonbox3]

Adding context for myself: https://learn.microsoft.com/en-us/rest/api/aifoundry/model-inference/get-chat-completions/get-chat-completions?view=rest-aifoundry-model-inference-2025-05-01&tabs=HTTP#chatcompletionsresponseformatjsonschema