LiteLLM Reasoning Agent Example

Use reasoning models through LiteLLM.

Use reasoning models through LiteLLM. The reasoning_content from the model response is extracted and displayed.

reasoning_agent.py

"""
LiteLLM Reasoning Agent Example

This example demonstrates using reasoning models through LiteLLM.
The reasoning_content from the model response is extracted and displayed.

Supported reasoning models through LiteLLM:
- deepseek/deepseek-reasoner (DeepSeek R1)
"""

from agno.agent import Agent
from agno.models.litellm import LiteLLM

# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------

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

# Using DeepSeek R1 through LiteLLM
agent = Agent(
    model=LiteLLM(
        id="deepseek/deepseek-reasoner",
    ),
    markdown=True,
)

agent.print_response(task, stream=True, stream_events=True, show_reasoning=True)

# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    pass

Run the Example

1. Set up your virtual environment

   Mac

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

   uv venv --python 3.12
   .venv\Scripts\activate
   
2. Install dependencies

   uv pip install -U agno litellm
   
3. Export your LiteLLM API key

   Mac/Linux

   export LITELLM_API_KEY="your_litellm_api_key_here"
   

   $Env:LITELLM_API_KEY="your_litellm_api_key_here"
   
4. Run the example

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

   python reasoning_agent.py
   

Full source: cookbook/90_models/litellm/reasoning_agent.py