Async Agent with Reasoning Capabilities

This example demonstrates the difference between a regular agent and a reasoning agent when solving mathematical problems, showcasing how reasoning mode provides more detailed thought processes.

Code

reasoning.py

import asyncio

from agno.agent import Agent
from agno.models.openai import OpenAIChat

task = "9.11 and 9.9 -- which is bigger?"

regular_agent = Agent(model=OpenAIChat(id="gpt-5-mini"), markdown=True)
reasoning_agent = Agent(
    model=OpenAIChat(id="gpt-5-mini"),
    reasoning=True,
    markdown=True,
)

asyncio.run(regular_agent.aprint_response(task, stream=True))
asyncio.run(
    reasoning_agent.aprint_response(task, stream=True, show_full_reasoning=True)
)

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 reasoning.py
   
5. Run Agent

   Mac

   python reasoning.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