xAI

Grok Imagine Video

The main Grok Imagine Video model can generate a clip from text, animate an image, incorporate reference images, edit a source video, or continue a video from its end. These modes use different inputs and constraints, while Grok Imagine Video 1.5 in Omnipix is the direct image-to-video path and should not be presented as a replacement for every workflow of the main model.

How to use this model

  1. 1

    Choose one workflow only: text generation, image animation, reference-guided generation, video editing, or video extension.

  2. 2

    Provide the matching source media and describe the scene, motion, camera, timing, and sound in the prompt.

  3. 3

    For reference generation, identify what each image contributes; for image-to-video, remember that the image becomes the opening frame.

  4. 4

    For an edit, state only the desired modifications and what must stay unchanged; for an extension, describe what happens after the source ends.

  5. 5

    Set duration, aspect ratio, and resolution only where the selected workflow exposes them, then review the complete temporal transition.

Imagine VideoAvailable

Imagine Video

The selected task determines the input contract: an image becomes the first frame, reference images guide content without fixing the first frame, a source video is modified in edit mode, and extension continues after the source ending. Do not combine inputs from different modes in one request.

Try now

Properties

  • Modes: text-to-video, image-to-video, reference-to-video, edit-video, and extend-video.
  • Generation controls include duration, aspect ratio, and resolution; edit mode inherits key properties from the source.

How to use this model

  1. 1

    Select exactly one task and attach only its matching media.

  2. 2

    Write a prompt covering action, camera, timing, and sound.

  3. 3

    Set duration, aspect ratio, and resolution where the task permits.

  4. 4

    For edits or extensions, state the change or continuation without redescribing preserved content unnecessarily.

Examples

Atmospheric landscape motion generated with Grok Imagine Video.

Parameters

Prompt

prompt
Optional

Task

task
Optional
Default: text_to_video

  • Text to Video

  • Image to Video

  • Reference to Video

  • Video Editing

  • Video Extension

Duration

duration
Optional
Range: 1 – 15

Aspect Ratio

aspect_ratio
Optional

  • 1:1

  • 16:9

  • 9:16

  • 4:3

  • 3:4

  • 3:2

  • 2:3

  • 2:1

  • 1:2

  • 19.5:9

  • 9:19.5

  • 20:9

  • 9:20

  • auto

Resolution

resolution
Optional

  • 480p

  • 720p

Image

image
Optional

Reference Images

reference_images
Optional

Video

video
Optional
Imagine Video 1.5Available

Imagine Video 1.5

Use 1.5 when the source image is the visual starting point and the task is straightforward image-to-video generation. It exposes motion prompting plus duration, aspect-ratio, and resolution controls, but not the main model's edit, extension, or multi-reference workflow.

Try now

Properties

  • Primary workflow: image-to-video from one supplied opening image.
  • Does not expose the main model's video edit, extension, or reference-image modes.

How to use this model

  1. 1

    Upload the opening image and describe the desired movement.

  2. 2

    Specify camera behavior and how the shot should develop over time.

  3. 3

    Choose duration, aspect ratio, and resolution for the output.

  4. 4

    Review identity, geometry, and the transition from the opening frame.

Parameters

Prompt

prompt
Required

Image

image
Required

Duration

duration
Required
Range: 1 – 15

Aspect Ratio

aspect_ratio
Required

  • 1:1

  • 16:9

  • 9:16

  • 4:3

  • 3:4

  • 3:2

  • 2:3

  • 2:1

  • 1:2

  • 19.5:9

  • 9:19.5

  • 20:9

  • 9:20

  • auto

Resolution

resolution
Required

  • 480p

  • 720p

  • 1080p