Skip to main content
POST
/
v1
/
tasks
Grok Imagine Video Series
curl --request POST \
  --url https://www.qingbo.dev/v1/tasks \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "action": "<string>",
  "prompt": "<string>",
  "aspect_ratio": "<string>",
  "resolution": "<string>",
  "duration": 123,
  "image_urls": [
    "<string>"
  ],
  "callback_url": "<string>",
  "callback_events": [
    "<string>"
  ]
}
'
{
  "task_id": "task-wave1775285160b950328499",
  "model": "grok-imagine-1.0-video",
  "action": "generate",
  "status": "queued",
  "created_at": 1775285160040,
  "progress": 0
}

Documentation Index

Fetch the complete documentation index at: https://docs.qingbo.dev/llms.txt

Use this file to discover all available pages before exploring further.

xAI Grok Imagine 1.0 video generation model — the video tier of the Grok multimodal family. Concise fields, low parameter overhead. Suited for social-media short videos, quick concept demos, and low-cost content pipelines.
The Grok Imagine series has two lines: an image tier and a video tier. For image generation see Grok Imagine Image; this page only covers the video tier.

​
Pricing

Billed per second by video duration, flat unit price across all resolutions: $0.0074375 / sec.
DurationCost
5 sec$0.037
10 sec$0.074
30 sec (max)$0.223

​
Examples

curl -X POST https://www.qingbo.dev/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-imagine-1.0-video",
    "prompt": "A rainy night street under neon lights, cyberpunk vibe, slow camera push-in",
    "duration": 6,
    "aspect_ratio": "16:9",
    "resolution": "480p"
  }'

{
  "task_id": "task-wave1775285160b950328499",
  "model": "grok-imagine-1.0-video",
  "action": "generate",
  "status": "queued",
  "created_at": 1775285160040,
  "progress": 0
}
After submission, poll status with GET /v1/tasks/{task_id}. See Task System for details.

​
Available Models

Model IDDescription
grok-imagine-1.0-videoxAI Grok Imagine 1.0 video tier, T2V / I2V, 5 aspect ratios, 480p / 720p, 6-30 seconds

​
Common Parameters

​
model
string
required
Fixed value grok-imagine-1.0-video
​
action
string
default:"generate"
Operation type. Allowed values:
  • generate — text-to-video (T2V)
  • image2video — image-to-video (I2V), used with image_urls
​
prompt
string
required
Video description. Required for T2V; in I2V mode, serves as motion / style guidance
​
aspect_ratio
string
default:"16:9"
Frame aspect ratio. Only effective in T2V; in I2V mode, automatically follows the reference image’s ratio. Allowed values:
  • 16:9 — landscape widescreen
  • 9:16 — portrait tall
  • 1:1 — square
  • 3:2 — landscape
  • 2:3 — portrait
​
resolution
string
default:"480p"
Output resolution. Allowed values:
  • 480p
  • 720p
​
duration
integer
default:"6"
Video duration in seconds, any integer in 6-30. Note: 30 seconds is the upper bound, which differs from most other video models (typically 5/10/15)
​
image_urls
string[]
Reference image URL array, up to 7 images. Passing this triggers I2V mode; the system automatically matches the reference image’s aspect ratio
​
callback_url
string
Webhook callback URL, invoked when the task reaches a terminal state. See Callbacks
​
callback_events
string[]
Callback event subscription list, by default only terminal-state events are pushed

​
Notes

  • Aspect ratio auto-matched in I2V mode — once image_urls is passed, the aspect_ratio field is ignored and the output ratio follows the first reference image.
  • Unusual duration range — 6-30 seconds, different from most video models in the catalog (which use the 5/10/15 tiers). When validating form input on the business side, remember to allow the full range.
  • Availability subject to upstream — upstream xAI / Grok quotas may fluctuate. For production, configure callback_url for async push instead of pure polling.

​
Resource Limits

ItemLimit
Reference imagesUp to 7, JPG/PNG/WEBP
Duration6-30 seconds
Resolution480p / 720p
OutputMP4, link valid for 24 hours