> ## Documentation Index
> Fetch the complete documentation index at: https://thethirdpenco-feat-tool-lifecycle-events.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# xAI

> Use xAI models (Grok) via the official API

xAI provides access to Grok logic and vision models through an OpenAI-compatible API.

## Setup

Set your API key as an environment variable:

```bash theme={null}
export XAI_API_KEY="xai-..."
```

Get your API key from [xAI Console](https://console.x.ai/).

## Usage

```python theme={null}
from ai_query import generate_text
from ai_query.providers import xai

result = await generate_text(
    model=xai("grok-2-1212"),
    prompt="Tell me a joke."
)
```

## Available Models

| Model ID             | Description                             |
| -------------------- | --------------------------------------- |
| `grok-2-1212`        | Grok 2 - Latest flagship model          |
| `grok-2-vision-1212` | Grok 2 Vision - Multimodal capabilities |
| `grok-beta`          | Grok Beta                               |
| `grok-vision-beta`   | Grok Vision Beta                        |

See the [xAI models documentation](https://docs.x.ai/docs/models) for a full list.

## Provider Options

Customize parameters:

```python theme={null}
result = await generate_text(
    model=xai("grok-2-1212"),
    prompt="Write a story.",
    provider_options={
        "xai": {
            "temperature": 0.9,
            "max_tokens": 1000
        }
    }
)
```

## Tool Calling

Tool calling is fully supported:

```python theme={null}
from ai_query import generate_text
from ai_query.providers import xai, tool, Field

@tool(description="Get weather")
async def get_weather(location: str = Field(description="City")) -> str:
    return f"Weather in {location}: 72°F, sunny"

result = await generate_text(
    model=xai("grok-2-1212"),
    prompt="What is the weather in SF?",
    tools={"get_weather": get_weather}
)
```

## Streaming

```python theme={null}
from ai_query import stream_text
from ai_query.providers import xai

result = stream_text(
    model=xai("grok-2-1212"),
    prompt="Write a poem."
)

async for chunk in result.text_stream:
    print(chunk, end="", flush=True)
```