Feature request: Add xAI Grok image-to-video generation support
Summary
Please add image-to-video support for the xAI (Grok) media provider in 9Router.
At the moment, Media Providers → Video → xAI (Grok) appears to support text-to-video generation only. There does not seem to be a way to provide an input image and generate a video based on that image.
Current behavior
The current xAI/Grok video integration can generate a video from a text prompt, but it does not expose an image-to-video workflow through the 9Router UI or API.
This limits use cases such as:
- Animating an existing image
- Bringing a generated character or illustration to life
- Creating a video while preserving the composition, subject, or style of a source image
Expected behavior
Please add support for image-to-video requests for the xAI/Grok video models, including:
- An image upload or image URL input in Media Providers → Video → xAI (Grok)
- Support for base64/data-URI images if that is required by the upstream API
- A text prompt describing the desired motion or transformation
- Any model-specific parameters supported by xAI, such as aspect ratio, duration, or resolution
- Polling and status handling for the resulting asynchronous video generation job
- A documented API format for clients that want to use image-to-video programmatically
If xAI uses a provider-specific request schema, 9Router could normalize it behind an OpenAI-compatible or otherwise consistent image-to-video interface, as appropriate for the existing media API design.
Example request shape
The exact schema can follow the upstream xAI API, but conceptually it could look like:
{
"model": "xai/grok-imagine-video",
"prompt": "Make the subject gently turn toward the camera and smile",
"image": {
"url": "https://example.com/input.jpg"
}
}Acceptance criteria
- Users can generate a video from an input image plus a text prompt through the 9Router UI.
- The corresponding API endpoint accepts an image URL and/or an uploaded/base64 image, according to xAI's supported input format.
- The generated video can be polled and retrieved using the normal 9Router flow.
- The supported Grok video models, input formats, and limitations are documented.
Related issue
Image-to-image support for Grok image models is already tracked in #1608. It would be useful to support both workflows so that the xAI/Grok media provider offers a more complete set of image and video capabilities.
Feature request: Add xAI Grok image-to-video generation support
Summary
Please add image-to-video support for the xAI (Grok) media provider in 9Router.
At the moment, Media Providers → Video → xAI (Grok) appears to support text-to-video generation only. There does not seem to be a way to provide an input image and generate a video based on that image.
Current behavior
The current xAI/Grok video integration can generate a video from a text prompt, but it does not expose an image-to-video workflow through the 9Router UI or API.
This limits use cases such as:
Expected behavior
Please add support for image-to-video requests for the xAI/Grok video models, including:
If xAI uses a provider-specific request schema, 9Router could normalize it behind an OpenAI-compatible or otherwise consistent image-to-video interface, as appropriate for the existing media API design.
Example request shape
The exact schema can follow the upstream xAI API, but conceptually it could look like:
{ "model": "xai/grok-imagine-video", "prompt": "Make the subject gently turn toward the camera and smile", "image": { "url": "https://example.com/input.jpg" } }Acceptance criteria
Related issue
Image-to-image support for Grok image models is already tracked in #1608. It would be useful to support both workflows so that the xAI/Grok media provider offers a more complete set of image and video capabilities.