This example demonstrates how it works when you pass a non-reasoning model as a reasoning model. It defaults to using the default OpenAI reasoning model. We recommend using the appropriate reasoning model or passing reasoning=True to use the default Chain-of-Thought reasoning.

​
Code

cookbook/reasoning/agents/default_chain_of_thought.py
from agno.agent import Agent
from agno.models.openai import OpenAIChat

reasoning_agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    reasoning_model=OpenAIChat(
        id="gpt-4o", max_tokens=1200
    ),  # Should default to manual COT because it is not a native reasoning model
    markdown=True,
)
reasoning_agent.print_response(
    "Give me steps to write a python script for fibonacci series",
    stream=True,
    show_full_reasoning=True,
)


# It uses the default model of the Agent
reasoning_agent = Agent(
    model=OpenAIChat(id="gpt-4o", max_tokens=1200),
    reasoning=True,
    markdown=True,
)
reasoning_agent.print_response(
    "Give me steps to write a python script for fibonacci series",
    stream=True,
    show_full_reasoning=True,
)

​
Usage

1

Create a virtual environment

Open the Terminal and create a python virtual environment.

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

Set your API key

export OPENAI_API_KEY=xxx
3

Install libraries

pip install -U openai agno
4

Run Example

python cookbook/reasoning/agents/default_chain_of_thought.py