Demo DeepSeek Qwen

Enable reasoning on Qwen 2.5 32B with DeepSeek R1 Distill Qwen as the reasoning model.

Groq retired the source’s Qwen-2.5-32b model in April 2025. Replace it with qwen/qwen3.6-27b. Groq retired the source’s Deepseek-r1-distill-qwen-32b model in April 2025. Replace it with qwen/qwen3.6-27b. See Groq deprecations.

demo_deepseek_qwen.py

"""Run `uv pip install ddgs sqlalchemy pgvector pypdf openai groq` to install dependencies."""

from agno.agent import Agent
from agno.models.groq import Groq

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

agent_with_reasoning = Agent(
    model=Groq(id="Qwen-2.5-32b"),
    reasoning=True,
    reasoning_model=Groq(
        id="Deepseek-r1-distill-qwen-32b", temperature=0.6, max_tokens=1024, top_p=0.95
    ),
)
agent_with_reasoning.print_response("9.11 and 9.9 -- which is bigger?", markdown=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 groq
   
3. Export your Groq API key

   Mac/Linux

   export GROQ_API_KEY="your_groq_api_key_here"
   

   $Env:GROQ_API_KEY="your_groq_api_key_here"
   
4. Replace Qwen 2.5 32B

   Replace Qwen-2.5-32b with qwen/qwen3.6-27b in the saved file.

5. Replace DeepSeek Qwen Distill

   Replace Deepseek-r1-distill-qwen-32b with qwen/qwen3.6-27b in the saved file.

6. Run the example

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

   python demo_deepseek_qwen.py
   

Full source: cookbook/90_models/groq/reasoning/demo_deepseek_qwen.py