# Llama OpenAI Image Input Bytes (/examples/models/meta/llama-openai/image-input-bytes)



<Warning>
  This source imports native `Llama` despite its OpenAI-compatible location. Apply the compatible adapter instructions below to use the endpoint described by this page. The dedicated `LlamaOpenAI` adapter has a formatter mismatch at the linked revision.
</Warning>

```python title="image_input_bytes.py"
"""
Meta Image Input Bytes
======================

Cookbook example for `meta/llama_openai/image_input_bytes.py`.
"""

from pathlib import Path

from agno.agent import Agent
from agno.media import Image
from agno.models.meta import Llama
from agno.tools.websearch import WebSearchTools
from agno.utils.media import download_image

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

agent = Agent(
    model=Llama(id="Llama-4-Maverick-17B-128E-Instruct-FP8"),
    tools=[WebSearchTools()],
    markdown=True,
)

image_path = Path(__file__).parent.joinpath("sample.jpg")

download_image(
    url="https://upload.wikimedia.org/wikipedia/commons/0/0c/GoldenGateBridge-001.jpg",
    output_path=str(image_path),
)

# Read the image file content as bytes
image_bytes = image_path.read_bytes()

agent.print_response(
    "Tell me about this image and give me the latest news about it.",
    images=[
        Image(content=image_bytes),
    ],
    stream=True,
)

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

if __name__ == "__main__":
    pass
```

## Run the Example [#run-the-example]

<Steps>
    <Step title="Set up your virtual environment">
      <CodeBlockTabs defaultValue="Mac">
        <CodeBlockTabsList>
          <CodeBlockTabsTrigger value="Mac">
            Mac
          </CodeBlockTabsTrigger>

          <CodeBlockTabsTrigger value="Windows">
            Windows
          </CodeBlockTabsTrigger>
        </CodeBlockTabsList>

        <CodeBlockTab value="Mac">
          ```bash
          uv venv --python 3.12
          source .venv/bin/activate
          ```
        </CodeBlockTab>

        <CodeBlockTab value="Windows">
          ```bash
          uv venv --python 3.12
          .venv\Scripts\activate
          ```
        </CodeBlockTab>
      </CodeBlockTabs>
    </Step>

  <Step title="Install dependencies">
    ```bash
    uv pip install -U agno ddgs llama-api-client openai
    ```
  </Step>

  <Step title="Export your Meta Llama API key">
    <CodeBlockTabs defaultValue="Mac/Linux">
      <CodeBlockTabsList>
        <CodeBlockTabsTrigger value="Mac/Linux">
          Mac/Linux
        </CodeBlockTabsTrigger>

        <CodeBlockTabsTrigger value="Windows">
          Windows
        </CodeBlockTabsTrigger>
      </CodeBlockTabsList>

      <CodeBlockTab value="Mac/Linux">
        ```bash
        export LLAMA_API_KEY="your_llama_api_key_here"
        ```
      </CodeBlockTab>

      <CodeBlockTab value="Windows">
        ```powershell
        $Env:LLAMA_API_KEY="your_llama_api_key_here"
        ```
      </CodeBlockTab>
    </CodeBlockTabs>
  </Step>

    <Step title="Use the compatible API adapter">
      Replace the `LlamaOpenAI` import (or `Llama` in the byte-image source) with this helper. Then replace every `LlamaOpenAI(...)` or `Llama(...)` construction in the saved file with `llama_model(...)`. Keep the existing `id`, `temperature`, and any retry options inside those calls.

      ```python title="Compatible model helper"
      from os import getenv

      from agno.models.openai.like import OpenAILike


      def llama_model(**kwargs):
          return OpenAILike(
              api_key=getenv("LLAMA_API_KEY"),
              base_url="https://api.llama.com/compat/v1/",
              supports_native_structured_outputs=False,
              supports_json_schema_outputs=True,
              **kwargs,
          )
      ```

      This uses Meta's OpenAI-compatible endpoint. You need a Meta API account with access to the selected model; check your account's current model catalog before running.
    </Step>

  <Step title="Run the example">
    Save the code above as `image_input_bytes.py`, then run:

    ```bash
    python image_input_bytes.py
    ```
  </Step>
</Steps>

Full source: [cookbook/90\_models/meta/llama\_openai/image\_input\_bytes.py](https://github.com/agno-agi/agno/blob/8f36eaf2d18e91afa7b327eec66a3cd3685dcb87/cookbook/90_models/meta/llama_openai/image_input_bytes.py)