How to deploy a Responses API on Azure AI Foundry?

Question

Hi

I would like to use REST with Responses API . Found the link https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/responses?tabs=rest-api

I created a new project, etc. at AI Foundry on the Azure portals, etc., but I see no option to deploy it. I deployed a gpt-4 model and then tried to the end point ending with responses as mentioned in the link above, but that didn't work.

I then thought that i should actually deploy the responses API but when i go to deployments in the project, i see no option to deploy responses api. I see all the other stuff like agents, chat, etc. but no responses APi!

What am i missing here? Am sure it is something silly!

Thanks in advance!

Answer 1

Hello It is VMS

Good day.

Response API is not available as a deployment icon in foundry but as an **API available with SDK or Rest API only. (client.responses.create instead of client.chat.completions.create)

gpt-4 is not supported for Responses API. Currently Response API is supported for below models

• gpt-4o (Versions: 2024-11-20, 2024-08-06, 2024-05-13)
• gpt-4o-mini (Version: 2024-07-18)
• computer-use-preview
• gpt-4.1 (Version: 2025-04-14)
• gpt-4.1-nano (Version: 2025-04-14)
• gpt-4.1-mini (Version: 2025-04-14)
• gpt-image-1 (Version: 2025-04-15)
• o3 (Version: 2025-04-16)
• o4-mini (Version: 2025-04-16)

With limited regions as mentioned in region availability section

Below is sample usage with Gpt-4.1-nano deploymet

import os
from openai import OpenAI
client = OpenAI(
    api_key=os.getenv("AZURE_OPENAI_API_KEY"),
    base_url="https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
    default_query={"api-version": "preview"}, 
)
response = client.responses.create(   
  model="gpt-4.1-nano", # Replace with your model deployment name 
  input="This is a test.",
)
print(response.model_dump_json(indent=2))

Hope you got the needed clarity now. Please let us know if you are still facing challenges

Thank you.