Gemini 3 Pro Thinking Level

Set Gemini 3.1 Pro Preview thinking level and stream an asynchronous response.

gemini_3_pro_thinking_level.py
"""
Async example using Gemini with tool calls.
"""

import asyncio

from agno.agent import Agent
from agno.models.google import Gemini

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

agent = Agent(
    model=Gemini(id="gemini-3.1-pro-preview", thinking_level="low"),
    markdown=True,
)

asyncio.run(agent.aprint_response("Whats the current news in France?", stream=True))

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

if __name__ == "__main__":
    pass

Run the Example

Set up your virtual environment

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

Install dependencies

uv pip install -U agno google-genai

Export your Google API key

export GOOGLE_API_KEY="your_google_api_key_here"

Choose a prompt for this configuration

The agent has no search tool. Replace the current-news prompt with a reasoning task, or enable search=True on Gemini when you need current web information.

Run the example

Save the code above as gemini_3_pro_thinking_level.py, then run:

python gemini_3_pro_thinking_level.py

Full source: cookbook/90_models/google/gemini/gemini_3_pro_thinking_level.py