When a request is deemed to violate xAI terms, generation may still be charged and credits are not refunded.
Drop files here, paste files, or add a URL. image/jpeg,image/png,image/webp · 10 MiB
Drop files here, paste files, or add a URL. image/jpeg,image/png,image/webp · 10 MiB
Not available when image_urls is provided.
Example Output
This is sample data. Generate a new result to see real output.
Example video
Transparent pricing with no hidden fees. Pay as you go.
grok-imagine-video-1.5grok-imagine-video-1.5grok-imagine-video-1.5grok-imagine-video-1.5| Model | Spec | PoYo price | Official price | You save |
|---|---|---|---|---|
| grok-imagine-video-1.5 | 480p | $0.072/sec 14.5 credits /sec | $0.080/sec Provider | 9% |
720p | $0.125/sec 25 credits /sec | $0.140/sec Provider | 11% | |
Input image | $0.010/image 2 credits /image | $0.010/image Provider | - | |
1080p - text/image mode | $0.225/sec 45 credits /sec | $0.250/sec Provider | 10% |
* Actual fees are based on the final output.
Complete guide to using Grok Imagine Video 1.5 API for Text, Image, and Reference Video
Three workflows in one API: text-to-video, first-frame image-to-video, and multi-reference video generation.
01
Start from one product frame and use the prompt to describe the motion you want. This sample follows a product-photography prompt for a brightly colored running shoe on mossy ground with a slow spin, keeping the object as the visual anchor while adding movement for ads, catalog pages, and launch concepts.
02
Grok Imagine Video 1.5 responds well to camera and scene cues such as tracking forward, rising shots, drifting mist, rippling water, and subject movement across frame. The fjord example uses those cues to turn a landscape still into a directed establishing shot.
03
When the scene needs atmosphere, include concise sound cues with the visual direction. This rain-town example pairs a character turn with falling rain and fishing-town ambience, so the prompt describes both the on-screen action and the intended audio mood.
Animate product shots into short clips for campaign drafts, catalog motion, and launch experiments.
Turn still concepts into moving scenes before committing to a full production workflow.
Generate short visual loops and motion variants from existing images for feeds, ads, and creator content.
Use a prepared frame as the visual anchor, then describe camera motion and subject behavior in the prompt.
Step 1: Create your API key
Generate a PoYo API key and prepare an asynchronous submit request.
Create API Key
Step 2: Provide a prompt and source image
Send model, prompt, and one image_urls item, then choose resolution and duration if you do not want the defaults.
Add Credits
Step 3: Poll status or use webhook
Use the returned task ID to check progress, or receive the final video through your callback URL.
Open API Documentation
Use one model ID and one submit endpoint for all three generation modes.
Resolution and duration are exposed directly, so production code can set cost and output targets per request.
The model page shows pricing by output tier and input image handling through the existing PoYo pricing table.
Integrate generated videos into backend workflows with task polling or callback-based completion handling.
