error using new foundry agent in streaming mode

[APiccardo4Quantis]

• Package Name:
  azure-ai-projects 2.0.0b2
  azure-core 1.36.0
• Operating System:
  w11
• Python Version:
  tryed 3.10 and 3.12.10

Describe the bug

hi!
trying to migrate to new version foundry agent. (docs not clear)
im using as an example the one given in this repo.

responses.create(..) works fine, but when trying to stream responses ,i cant get it to work

this is my code:

openai_client = project_client.get_openai_client()
agent_model= get_agent_model(agent_name) #this just return the model config in foundry
print(agent_model)

# Optional Step: Create a conversation to use with the agent
conversation = openai_client.conversations.create()
print(f"Created conversation (id: {conversation.id})")

# Chat with the agent to answer questions
with openai_client.responses.stream(
    conversation=conversation.id,
    #model=agent_model,
    extra_body={"agent": {"name": agent_name, "type": "agent_reference"}},
    input="hi!",
) as stream:
    for event in stream:
        if event.type == "response.output_text.delta":    
            print(event.delta, end="", flush=True)

---

versions:
azure-ai-agents 1.1.0
azure-ai-projects 2.0.0b2
azure-core 1.35.1

console error:
Created conversation (id: conv_23ee7a807b7307ce00nGBsefUEqRE448CvhIyABlMEqNJOPqoQ)
Traceback (most recent call last):
File "c:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\dev\scripts\NewApi.py", line 36, in
with openai_client.responses.stream(
File "C:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\env\lib\site-packages\openai\resources\responses\responses.py", line 1005, in stream
raise ValueError("model must be provided when creating a new response")
ValueError: model must be provided when creating a new response

so the first thing i did was:
uncomment the model line and this error happend:

console error:
Created conversation (id: conv_24f10f51dc2a61b900s2HxndrRWcRPtFYEZOmp3iyODsSbiSgE)
Traceback (most recent call last):
File "c:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\dev\scripts\NewApi.py", line 36, in
with openai_client.responses.stream(
File "C:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\env\lib\site-packages\openai\lib\streaming\responses_responses.py", line 111, in enter
raw_stream = self.__api_request()
File "C:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\env\lib\site-packages\openai\resources\responses\responses.py", line 828, in create
return self._post(
File "C:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\env\lib\site-packages\openai_base_client.py", line 1259, in post
return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
File "C:\Users\Alejandro Piccardo\Desktop\cambios quantis\agente\BackendFastapi-Agent\env\lib\site-packages\openai_base_client.py", line 1047, in request
raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'code': 'invalid_payload', 'message': 'Not allowed when agent is specified.', 'param': 'model', 'type': 'invalid_request_error', 'details': [], 'additionalInfo': {'request_id': '7153967e656a11df001f063de6275551'}}}

what im doing wrong? or streams are not implemented correctly for agents??? can you guys provide me with an example response.stream(..)

thanks.

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @dargilco @glharper @howieleung @kingernupur @nick863 @trangevi @trrwilson.

[dargilco]

@APiccardo4Quantis for now please use the .response.create(stream=Ture, ...) method instead, per the sample code below. I'll add it to our samples folder.. I believe the .responses.stream() method is not yet supported.

# ------------------------------------
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
# ------------------------------------

"""
DESCRIPTION:
    This sample demonstrates how to run basic Prompt Agent operations
    using the synchronous AIProjectClient and OpenAI clients. The response
    is streamed.

    The OpenAI compatible Responses and Conversation calls in this sample are made using
    the OpenAI client from the `openai` package. See https://platform.openai.com/docs/api-reference
    for more information.

USAGE:
    python sample_agent_stream_events.py

    Before running the sample:

    pip install "azure-ai-projects>=2.0.0b1" python-dotenv

    Set these environment variables with your own values:
    1) AZURE_AI_PROJECT_ENDPOINT - The Azure AI Project endpoint, as found in the Overview
       page of your Microsoft Foundry portal.
    2) AZURE_AI_MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in
       the "Models + endpoints" tab in your Microsoft Foundry project.
"""

import os
from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient
from azure.ai.projects.models import PromptAgentDefinition

load_dotenv()

endpoint = os.environ["AZURE_AI_PROJECT_ENDPOINT"]

with (
    DefaultAzureCredential() as credential,
    AIProjectClient(endpoint=endpoint, credential=credential) as project_client,
    project_client.get_openai_client() as openai_client,
):

    agent = project_client.agents.create_version(
        agent_name="MyAgent",
        definition=PromptAgentDefinition(
            model=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
            instructions="You are a helpful assistant that answers general questions",
        ),
    )
    print(f"Agent created (id: {agent.id}, name: {agent.name}, version: {agent.version})")

    conversation = openai_client.conversations.create(
        items=[{"type": "message", "role": "user", "content": "Give me 5 good reasons why I should exercise daily."}],
    )
    print(f"Created conversation with initial user message (id: {conversation.id})")

    with openai_client.responses.create(
        conversation=conversation.id,
        extra_body={"agent": {"name": agent.name, "type": "agent_reference"}},
        input="",
        stream=True,
    ) as response_stream_events:

        for event in response_stream_events:
            if event.type == "response.created":
                print(f"Stream response created with ID: {event.response.id}\n")
            elif event.type == "response.output_text.delta":
                print(event.delta, end="", flush=True)
            elif event.type == "response.text.done":
                print(f"\nResponse text done. Access final text in 'event.text'")
            elif event.type == "response.completed":
                print(f"\nResponse completed. Access final text in 'event.response.output_text'")


    openai_client.conversations.delete(conversation_id=conversation.id)
    print("Conversation deleted")

    project_client.agents.delete_version(agent_name=agent.name, agent_version=agent.version)
    print("Agent deleted")

[github-actions]

Hi @APiccardo4Quantis. Thank you for opening this issue and giving us the opportunity to assist. We believe that this has been addressed. If you feel that further discussion is needed, please add a comment with the text "/unresolve" to remove the "issue-addressed" label and continue the conversation.