Does Grok Imagine text-to-video require an image?

No. grok-imagine/text-to-video accepts text only; use its image-to-video route for a source frame.

Which formats does Grok Imagine support?

It supports 6–30 seconds at 480p or 720p in 2:3, 3:2, 1:1, 16:9, or 9:16.

What are normal and fun modes?

normal targets general motion and steadier shots, while fun is for more energetic creative motion. Start with normal to validate subject and composition.

How is Grok Imagine priced?

Billing is per output second, with separate 480p and 720p rates.

How do I call the Grok Imagine API?

POST to /v1/tasks with model set to grok-imagine/text-to-video and place the generation fields inside input.

Back to Models

Grok Imagine Text to Video

by xAIVideo

xAI Grok Imagine text-to-video: generate short cinematic clips from a text prompt with selectable motion mode, aspect ratio, duration (6-30s), and 480p/720p resolution.

Pricing

Standard Usage

30 Credits/ s

Input

Prompt *0 / 5000

Duration6s

6s1-second steps30s

Choose 6-30 seconds in 1-second steps

Resolution

Aspect Ratio

More parameters

Generation settings

Mode

Estimated cost180 Credits30 Credits / s × 6s × 480p

Output

Ready for generation

Configure your parameters and click "Run" to see the output here.

Public model preview

From one prompt to a complete generated video

Three public API results cover science fiction, organic motion, and urban action. Play each clip and copy the prompt to test a variation.

Zero-gravity motion6s480p16:9

Rotate an astronaut inside a neon space station

Tests body rotation, floating motion, reflective light, and a readable science-fiction environment.

Prompt shown with this public preview

宇航员在霓虹空间站内失重旋转

Organic motion6s480p16:9

Move through a bioluminescent deep sea

Uses drifting jellyfish and soft light to test fluid, low-gravity movement.

Prompt shown with this public preview

深海水母在生物荧光中优雅游动

Urban action6s480p16:9

Race through wet streets at night

Combines vehicle speed, rain, reflections, and dense city lighting in one short clip.

Prompt shown with this public preview

复古赛车雨夜疾驰穿过城市街道

Choose by input method, output specifications, control surface, and budget.

About Grok Imagine Text to Video

Grok Imagine turns text into 6–30 second short videos at 480p or 720p. It offers normal and fun motion modes for general scenes, social clips, and fast ideation.

This is a pure text-to-video route with no source-image field. Choose the separate image-to-video model when an exact opening frame is required. grok-imagine/text-to-video

Provider: xAI
Task: Pure text-to-video
Input: Text prompt + motion mode
Starting price: 30 Credits / second

Model specifications

These specifications reflect the current HiAPI Grok Imagine Text to Video request contract.

Input: Text prompt + motion mode
Duration: 6–30s
Resolution: 480p720p
Aspect ratio: 2:33:21:116:99:16
Audio and references: No audio control field
API protocol: POST /v1/tasksAsync

480p： 30 Credits / second720p： 56 Credits / secondBilled per generated second by resolution

How to use Grok Imagine Text to Video

Start at 480p with the normal mode, then test fun mode or 720p after motion and framing are stable.

Subject

Action

Camera

Define subject, action, and camera

Grok Imagine Text to Video API quickstart

This is the minimum request for an asynchronous text-to-video task. Use the API docs for task retrieval, callbacks, errors, and the complete parameter contract.

curl -X POST "https://api.hiapi.ai/v1/tasks" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "grok-imagine/text-to-video",
  "input": {
    "prompt": "A cyberpunk street under neon lights, rain reflecting the signs, slow push-in",
    "duration": 6,
    "resolution": "480p",
    "aspect_ratio": "16:9",
    "mode": "normal"
  }
}'

The create request returns a taskId; retrieve the result asynchronously.Open full API docs