Grok Imagine Video - Unifically

Generate high-quality AI videos using xAI’s Grok Imagine model.

xAI upgraded the model behind xai/grok-imagine-video: it used to run Grok Imagine 1.0 Video, and now runs a speed-optimized build of Grok Imagine 1.5 Video — faster results in exchange for a small amount of quality. This build differs from the official API version, which is available separately as Grok Imagine Video 1.5.

Model

xai/grok-imagine-video

Parameters

Parameter	Type	Required	Default	Allowed Values	Description
`prompt`	string	Yes	—	—	Video generation prompt. Use `@ImageN` to point at a specific reference from `image_urls`
`aspect_ratio`	string	No	`"1:1"`	`"1:1"`, `"2:3"`, `"3:2"`, `"9:16"`, `"16:9"`	Output aspect ratio
`duration`	int	No	`6`	`1`–`30`	Video duration in seconds
`resolution`	string	No	`"720p"`	`"480p"`, `"720p"`	Output resolution
`start_image_url`	string	No	`null`	—	Start frame image URL. Cannot be combined with `image_urls`
`image_urls`	string[]	No	`null`	1–7 URLs	Reference image URLs for style, subject, or scene guidance. Cannot be combined with `start_image_url`
`video_preset`	string	No	`"custom"`	`"custom"`, `"spicy"`, `"fun"`, `"normal"`	Generation style preset

start_image_url and image_urls are mutually exclusive — send one or the other, never both. image_urls accepts a maximum of 7 reference images.

Video presets:

Preset	Style
`custom`	Default generation
`spicy`	Extremely spicy/nsfw
`fun`	Extremely crazy/fun
`normal`	Normal style

Example Request

Text to Video

curl -X POST https://api.unifically.com/v1/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "xai/grok-imagine-video",
    "input": {
      "prompt": "A futuristic cityscape at sunset with flying cars",
      "duration": 10,
      "resolution": "720p",
      "aspect_ratio": "16:9",
      "video_preset": "custom"
    }
  }'

Image to Video

curl -X POST https://api.unifically.com/v1/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "xai/grok-imagine-video",
    "input": {
      "prompt": "Camera slowly zooms in while clouds move across the sky",
      "start_image_url": "https://example.com/landscape.jpg",
      "duration": 6,
      "resolution": "720p",
      "aspect_ratio": "16:9"
    }
  }'

Text to Video with Reference Images

curl -X POST https://api.unifically.com/v1/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "xai/grok-imagine-video",
    "input": {
      "prompt": "The woman in @Image1 walks into the neon-lit bar from @Image2 and sits down at the counter, camera tracking her from behind",
      "image_urls": [
        "https://example.com/woman.jpg",
        "https://example.com/neon-bar.jpg"
      ],
      "duration": 10,
      "resolution": "720p",
      "aspect_ratio": "16:9"
    }
  }'

Prompt References

Token	Refers to
`@Image1`, `@Image2`, …	Items from `image_urls`, in the order they are sent

Reference the assets you pass in image_urls directly from the prompt, so you can assign each image a role instead of describing it in words. With a character image first and a location image second:

The woman in @Image1 walks into the neon-lit bar from @Image2 and sits
down at the counter, camera tracking her from behind

Prompt references are only available on xai/grok-imagine-video (the speed-optimized 1.5 build). Grok Imagine Video 1.5 does not support them — it takes a single start_image_url and no reference assets.

Response

{
  "code": 200,
  "success": true,
  "data": {
    "task_id": "abc123def456",
    "status": "processing"
  }
}

​Model

​Parameters

​Example Request

​Text to Video

​Image to Video

​Text to Video with Reference Images

​Prompt References

​Response

Model

Parameters

Example Request

Text to Video

Image to Video

Text to Video with Reference Images

Prompt References

Response