grok-imagine-1.0-video

Grok Imagine text/img-to-video, discrete durations.

Modes

text→video · img→video

Size

720x1280 / 1280x720 / 1024x1024 / 1792x1024 / 1024x1792

Seconds

discrete 6 / 10 / 12 / 16

Refs

up to 4, optional

Billing

per second

Model notes

• Duration is discrete: only 6 / 10 / 12 / 16 seconds.
• 5 fixed sizes: 720x1280 / 1280x720 / 1024x1024 / 1792x1024 / 1024x1792.
• Text-only or image-to-video; up to 4 reference images.
• With 2 or more reference images, duration is capped at 10 seconds (only 6 / 10).

Request example

{
  "model": "grok-imagine-1.0-video",
  "prompt": "a neon fox running through a city at night",
  "size": "1280x720",
  "seconds": 10
}