> ## 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.

# Meta

> Use Meta Llama models with Agno agents.

Meta offers multi-modal language models with strong text understanding and visual intelligence.

We recommend experimenting to find the best-suited model for your use-case. Here are some general recommendations:

* `Llama-4-Scout-17B-16E-Instruct-FP8`: Excellent performance for most general tasks, including multi-modal scenarios.
* `Llama-3.3-70B-Instruct`: Strong instruction-following model for complex reasoning tasks.

Explore all the models [here](https://llama.developer.meta.com/docs/models).

## Installation

```bash theme={null}
uv pip install -U llama-api-client agno
```

## Authentication

Set your `LLAMA_API_KEY` environment variable:

<CodeGroup>
  ```bash Mac theme={null}
  export LLAMA_API_KEY=YOUR_API_KEY
  ```

  ```bash Windows theme={null}
  setx LLAMA_API_KEY YOUR_API_KEY
  ```
</CodeGroup>

## Example

Use `Llama` with your `Agent`:

<CodeGroup>
  ```python agent.py theme={null}
  from agno.agent import Agent
  from agno.models.meta import Llama

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

  agent.print_response("Share a 2 sentence horror story.")

  ```
</CodeGroup>

<Note> View more examples [here](/models/providers/native/meta/usage/basic). </Note>

## Parameters

| Parameter  | Type                              | Default                                    | Description                                                         |
| ---------- | --------------------------------- | ------------------------------------------ | ------------------------------------------------------------------- |
| `id`       | `str`                             | `"Llama-4-Maverick-17B-128E-Instruct-FP8"` | The id of the Meta model to use                                     |
| `name`     | `str`                             | `"Llama"`                                  | The name of the model                                               |
| `provider` | `str`                             | `"Llama"`                                  | The provider of the model                                           |
| `api_key`  | `Optional[str]`                   | `None`                                     | The API key for the Llama API (defaults to `LLAMA_API_KEY` env var) |
| `base_url` | `Optional[Union[str, httpx.URL]]` | `None`                                     | The base URL for the Llama API (defaults to the Llama SDK default)  |

### OpenAI-like Parameters

Agno also provides `LlamaOpenAI`, a client that calls the Llama API through its OpenAI-compatible endpoint (`https://api.llama.com/compat/v1/`). It uses the `openai` package. Install it with `uv pip install -U openai`. Use it in place of `Llama` when you want OpenAI-style request handling:

```python theme={null}
from agno.models.meta import LlamaOpenAI

agent = Agent(model=LlamaOpenAI(id="Llama-4-Maverick-17B-128E-Instruct-FP8"))
```

`LlamaOpenAI` supports all parameters from [OpenAI Like](/reference/models/openai-like).

## Resources

* [Meta AI Models](https://llama.developer.meta.com/docs/models)
* [Llama API Documentation](https://llama.developer.meta.com/docs/overview)