> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agno.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Image Generation Agent (Streaming)

> Stream image generation from a Gemini agent with response_modalities.

## Code

```python image_generation_stream.py theme={null}
from io import BytesIO

from agno.agent import Agent, RunOutput  # noqa
from agno.models.google import Gemini
from PIL import Image

# No system message should be provided
agent = Agent(
    model=Gemini(
        id="gemini-3.1-flash-image",
        response_modalities=["Text", "Image"],
    )
)

# Print the response in the terminal
response = agent.run("Make me an image of a cat in a tree.", stream=True)

for chunk in response:
    if hasattr(chunk, "image") and chunk.image:  # type: ignore
        image_bytes = chunk.image.content  # type: ignore
        if image_bytes:
            image = Image.open(BytesIO(image_bytes))
            image.show()
            # Save the image to a file
            # image.save("generated_image.png")
```

## Usage

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Set your API key">
    ```bash theme={null}
    export GOOGLE_API_KEY=xxx
    ```
  </Step>

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U google-genai pillow agno
    ```
  </Step>

  <Step title="Run Agent">
    ```bash theme={null}
    python image_generation_stream.py
    ```
  </Step>
</Steps>