Async Agent Streaming Responses

This example demonstrates different methods of handling streaming responses from async agents, including manual iteration, print response, and pretty print response.

Code

streaming.py

import asyncio

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.utils.pprint import apprint_run_response

agent = Agent(
    model=OpenAIChat(id="gpt-5-mini"),
)


async def streaming():
    async for response in agent.arun(input="Tell me a joke.", stream=True):
        print(response.content, end="", flush=True)


async def streaming_print():
    await agent.aprint_response(input="Tell me a joke.", stream=True)


async def streaming_pprint():
    await apprint_run_response(agent.arun(input="Tell me a joke.", stream=True))


if __name__ == "__main__":
    asyncio.run(streaming())
    # OR
    asyncio.run(streaming_print())
    # OR
    asyncio.run(streaming_pprint())

Usage

1. Create a virtual environment

   Open the Terminal and create a python virtual environment.

   Mac

   python3 -m venv .venv
   source .venv/bin/activate
   
2. Install libraries

   pip install -U agno openai
   
3. Export your OpenAI API key

   Mac/Linux

     export OPENAI_API_KEY="your_openai_api_key_here"
   
4. Create a Python file

   Create a Python file and add the above code.

   touch streaming.py
   
5. Run Agent

   Mac

   python streaming.py
   
6. Find All Cookbooks

   Explore all the available cookbooks in the Agno repository. Click the link below to view the code on GitHub:

   Agno Cookbooks on GitHub