Azure OpenAI Basic Reasoning Stream

Adapt the archived GPT-4.1 stage to a native Azure OpenAI reasoning deployment.

basic_reasoning_stream.py
"""
Basic Reasoning Stream
======================

Demonstrates this reasoning cookbook example.
"""

import asyncio

from agno.agent import Agent
from agno.models.azure.openai_chat import AzureOpenAI
from agno.run.agent import RunEvent  # noqa


# ---------------------------------------------------------------------------
# Create Example
# ---------------------------------------------------------------------------
def run_example() -> None:
    async def streaming_reasoning():
        """Test streaming reasoning with a Azure OpenAI model."""
        # Create an agent with reasoning enabled
        agent = Agent(
            reasoning_model=AzureOpenAI(id="gpt-4.1"),
            instructions="Think step by step about the problem.",
        )

        prompt = "What is 25 * 37? Show your reasoning."

        await agent.aprint_response(prompt, stream=True, stream_events=True)

        # Use manual event loop to see all events
        # async for run_output_event in agent.arun(
        #     prompt,
        #     stream=True,
        #     stream_events=True,
        # ):
        #     if run_output_event.event == RunEvent.run_started:
        #         print(f"\nEVENT: {run_output_event.event}")

        #     elif run_output_event.event == RunEvent.reasoning_started:
        #         print(f"\nEVENT: {run_output_event.event}")
        #         print("Reasoning started...\n")

        #     elif run_output_event.event == RunEvent.reasoning_content_delta:
        #         # This is the NEW streaming event for reasoning content
        #         print(run_output_event.reasoning_content, end="", flush=True)

        #     elif run_output_event.event == RunEvent.reasoning_step:
        #         print(f"\nEVENT: {run_output_event.event}")

        #     elif run_output_event.event == RunEvent.reasoning_completed:
        #         print(f"\n\nEVENT: {run_output_event.event}")

        #     elif run_output_event.event == RunEvent.run_content:
        #         if run_output_event.content:
        #             print(run_output_event.content, end="", flush=True)

        #     elif run_output_event.event == RunEvent.run_completed:
        #         print(f"\n\nEVENT: {run_output_event.event}")

    if __name__ == "__main__":
        asyncio.run(streaming_reasoning())


# ---------------------------------------------------------------------------
# Run Example
# ---------------------------------------------------------------------------
if __name__ == "__main__":
    run_example()

Select a native reasoning deployment

The archived AzureOpenAI(id="gpt-4.1") is not recognized as a native reasoning model. Agno can skip that stage and still produce a final answer. Deploy GPT-5.2 in your Azure resource, then replace the reasoning_model value with:

AzureOpenAI(id="gpt-5.2", azure_deployment="your-reasoning-deployment")

Replace the deployment placeholder with the name from your resource. Keep the underlying model ID in id so Agno can recognize its reasoning capability. AZURE_OPENAI_API_KEY and AZURE_OPENAI_ENDPOINT identify the resource; they do not create the deployment.

An explicit reasoning_model runs as a separate, tool-free reasoning stage before the main model response. show_full_reasoning=True displays the reasoning data Agno receives; it cannot reveal a provider's private internal trace. Some adapters use the reasoning stage's answer text when separate reasoning content is unavailable. A failed reasoning stage can still be followed by a main-model answer, so a completed run alone does not prove the reasoning stage succeeded.

This program omits the final model, so the final answer uses Agno's default OpenAI model and still needs OPENAI_API_KEY separately from the Azure credentials.

Run the Example

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Install dependencies

uv pip install -U agno openai

Export environment variables

export AZURE_OPENAI_API_KEY="your_azure_openai_api_key_here"
export AZURE_OPENAI_ENDPOINT="your_azure_openai_endpoint_here"
export OPENAI_API_KEY="your_openai_api_key_here"

Run the example

Save the code above as basic_reasoning_stream.py, then run:

python basic_reasoning_stream.py

Full source: cookbook/10_reasoning/models/azure_openai/basic_reasoning_stream.py