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

# xAI

> Image and video generation from xAI. Another model to reach for when a page needs a different look.

Image and video generation from xAI. Another model to reach for when a page needs a different look.

## Connecting

In the dashboard, open **Connectors**, find xAI and press **Connect**. The key you enter is checked against xAI before anything is saved, so a key that is refused never becomes a connection that fails later.

### What you will need

| Field   | What it is                         |
| ------- | ---------------------------------- |
| API Key | Your xAI API key from console.x.ai |

### Getting the credential

1. Sign up at [console.x.ai](https://console.x.ai)
2. Create an API key under **API Keys**
3. Paste it into the connector

## Notes

* **Video is asynchronous.** `generate_video` returns a `request_id`; call `get_video` with it until `status`
  is `done`. A video can also come back `failed` or `expired`.
* Both image and video results are copied onto BoostGPT storage. xAI serves them from temporary URLs that stop
  working, so the link the agent hands back is ours, not theirs.
* xAI does not offer a speech API. The previous `generate_speech` tool called an endpoint that answers 403 for
  every account and has been removed.

## What an agent can do with it

| Tool             | What it does                                                                                                                                                                                 |
| ---------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `generate_image` | Generate an image from a text prompt with Grok Imagine. The image is rehosted on a durable URL — xAI's own link is temporary.                                                                |
| `generate_video` | Start a Grok Imagine video generation from a prompt, optionally animating a still image. Asynchronous: returns a request\_id — poll it with get\_video.                                      |
| `get_video`      | Poll a video started by generate\_video. While it runs the status is 'pending'; when it is 'done' the finished video is rehosted on a durable URL. Status can also be 'failed' or 'expired'. |

## Useful resources

* [xAI API docs](https://docs.x.ai/api)
* [Grok Imagine](https://docs.x.ai/developers/model-capabilities/imagine)