Grok

Grok Video API

Video APIHOT

Grok Video API gives teams an API-first way to use Grok Imagine video generation inside production workflows. It supports prompt-driven text-to-video, reference-based image-to-video, video upscaling, and video extension from one model family.

Use the playground to validate prompts, source images, task IDs, duration, aspect ratio, and payload behavior before connecting Grok video generation to your app, creative tool, or content pipeline.

Model Type:
Pricing

Lowest cost starts at $0.008/s

Selected model
Grok Imagine Video 1.5 Preview

480p: 1.6 credits/s - Text/Image to Video - Output seconds × 1.6

720p: 3 credits/s - Text/Image to Video - Output seconds × 3

USD estimates use $0.005/credit with the best annual plan. Final billing follows the selected Grok video variant and output settings.

New tasks appear here after you submit.
API Input

Core request fields

Use a predictable payload for prompts, image URLs, source task IDs, aspect ratio, and callback settings.

Request JSON
{
  "model": "grok-imagine-video-1-5-preview",
  "input": {
    "prompt": "A cinematic product reveal with smooth orbit camera movement",
    "reference_image_urls": [
      "https://example.com/product.png"
    ],
    "reference_video_urls": [],
    "reference_audio_urls": [],
    "generate_audio": true,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "duration": 5
  }
}

Use Grok Imagine Video 1.5 Preview for generation

The playground, request sample, request fields, and pricing on this page follow the selected model type.

View detailed API docs

Open the full API documentation for endpoints, authentication, task polling, and payload details. /models/grok-imagine.md

promptstringPrompt for video generation. Maximum length: 4096 characters.
image_urlsarrayUpload image files to be used as API input. Supported file types: image/jpeg, image/png, image/webp, image/jpg. Maximum file size: 20MB. Supports multi-file upload, up to 7 file.Only one is supported when the resolution is 1080p.
aspect_ratioenumThe aspect ratio of the video. This parameter is invalid if it is a single image. (1:1, 16:9, 9:16, 3:2, 2:3, auto)
resolutionenumResolution for video generation. (480p, 720p, 1080p)
durationnumberVideo duration in seconds. Range: [1, 15]. Default: 8. Minimum: 1. Maximum: 15. Step: 1.
nsfw_checkerbooleanDefaults to false. You can set it to false based on your needs. If set to false, our content filtering will be disabled, and all results will be returned directly by the model itself.
Capabilities

What Grok Video API adds to your video workflow

The page follows the same API-first structure as the reference: test the model, understand the fields, then move into integration.

Text-to-video generation

Turn detailed prompts into short-form video concepts for social posts, ads, product moments, and fast creative exploration.

Image-to-video motion

Animate source images while preserving the subject, composition, and visual direction that your team already approved.

Upscale and extend

Improve successful Grok clips with video upscaling, or continue an existing generation with an extension task.

API-ready payloads

Work with model IDs, prompts, image URLs, task IDs, aspect ratios, callbacks, and task polling in one integration path.

Quick Start

From Grok video test to production request

A compact path for teams that want to validate output quality before wiring Grok video generation into an app.

1. Choose a mode

Select text-to-video for prompt-only creation, image-to-video for source image animation, or a follow-up mode for upscale and extend.

2. Submit generation

Call the API with your account key, model ID, prompt, optional image URLs or source task ID, and callback URL.

3. Poll and reuse

Track the task until it succeeds, then reuse the output URL in your product, campaign workflow, or follow-up video task.

Use Cases

What you can create with Grok Video API

Use the API when your product needs repeatable Grok video output, not just one-off manual generations.

Prompt-led social clips for TikTok, Reels, and Shorts
Animated product stills and campaign visuals
Rapid storyboard and concept video exploration
Video variants for ads, thumbnails, and landing pages
Upscaled versions of successful generated clips
Extended clips for iterative creative pipelines
Community Showcase

Successful Grok video generations from the community

Browse public Grok video outputs that finished successfully, then open any case to inspect the prompt, settings, and result.

View more
videoGrok Imagine Image to Video

Grok Imagine Image to Video

The woman is very shy. She shows off her femininity and beauty in front of the camera without speaking. I want soft romantic music playing in the background. I want a static shot—meaning a scene filmed from a single, fixed angle.

Sep 14, 2026View details
videoGrok Imagine Image to Video

Grok Imagine Image to Video

A woman shows off her femininity and beauty in front of the camera.

Sep 14, 2026View details
videoGrok Imagine Image to Video

Grok Imagine Image to Video

woman dances sensetive

Sep 14, 2026View details
videoGrok Imagine Image to Video

Grok Imagine Image to Video

Create a realistic UGC-style video. A person is holding the tablet and slowly moves it toward the camera. Natural smartphone footage, realistic home environment, handheld camera, natural lighting. No dialogue. No subtitles. No text. No scene changes.

Sep 13, 2026View details
videoGrok Imagine Text to Video

Grok Imagine Text to Video

27-year-old adult woman, photorealistic, Mediterranean European appearance, oval symmetrical face, defined cheekbones, soft jawline, full natural lips, small beauty mark under left cheekbone, almond-shaped hazel-green eyes, long dark eyelashes, naturally arched eyebrows, long wavy chocolate-brown hair with subtle caramel highlights, lightly tanned skin, subtle freckles across nose, natural realistic skin texture, feminine hourglass figure, slim waist, long legs, athletic toned physique, elegant, confident, sophisticated, luxury influencer aesthetic, natural makeup, nude lips, subtle eyeliner, minimalist fashion, high-end editorial photography, realistic proportions, natural lighting, DSLR photo, 85mm lens, shallow depth of field, highly detailed, consistent facial identity

Sep 4, 2026View details
videoGrok Imagine Image to Video

Grok Imagine Image to Video

Classic blocky R6 Roblox avatar animation rendered in Blender, 1:2:2 head-to-torso-to-legs proportions, cuboid torso, two one-piece straight rectangular prism arms visibly joined at the shoulders with no elbows or segmented forearms, two one-piece rectangular legs, cuboid head with no neck, flat 2D cartoon face decal, semi-realistic detailed environment, soft global illumination, shallow depth of field, vertical 9:16, no text, no captions, no labels, no character name tags, no watermark. Dax has default bright yellow Roblox skin, black spiky hair mesh, solid red tee torso, solid black pants, no accessories, and a flat 2D devastated recognition decal with huge watery eyes, one sweat drop, and a tiny fixed frown. Milo appears softly out of focus behind him with default bright yellow Roblox skin, bald/no hair beneath an angular silver crown accessory, solid lime-green tee torso, solid blue pants, no other accessories, and a flat 2D stern neutral decal. The tightest extreme close-up of the

Sep 4, 2026View details
Production Notes

Built for real Grok video API workflows

Account-gated execution

Users sign in, submit jobs, and consume credits through the existing SdanceAI generation flow.

Reference and task inputs

Image-to-video accepts source image URLs, while upscale and extend workflows can continue from completed video task IDs.

Mode-aware iteration

Start with text or image generation, then move promising outputs into upscale or extension workflows when needed.

FAQ

Grok Video API questions

What is Grok Video API best for?

It is best for repeatable short-form video generation, including prompt-led clips, animated source images, video variants, upscaled outputs, and extended creative sequences.

Which Grok video modes are available?

The current page exposes Grok Imagine Text to Video, Grok Imagine Image to Video, Grok Imagine Video Upscale, and Grok Imagine Video Extend.

What is the difference between Text to Video and Image to Video?

Text to Video starts from a prompt only. Image to Video uses one or more source images plus a prompt, which is better when you need subject or composition continuity.

When should I use upscale or extend?

Use upscale when you already have a successful Grok video and want a higher-quality version. Use extend when you want to continue a completed generation into a longer sequence.

Where can I test it before integrating?

Use the playground on this page to test prompts, image URLs, source task IDs, aspect ratios, callbacks, and API payload behavior before moving into production.

Does the API support callbacks?

Yes. The request schema includes callback settings where supported, so your service can receive task completion notifications.

Can I use Grok Video API for commercial workflows?

This page is configured for production video workflows. For commercial use, review your account plan, credit balance, model availability, and applicable content policy requirements.

Can I animate an existing product image?

Yes. Use the image-to-video mode with a source image URL and a prompt that describes the motion, camera behavior, and final visual direction.

How should I manage cost?

Start with quick prompt or image-to-video tests, then reserve upscale and extension tasks for clips that are already close to your target.

Can I inspect the actual API fields?

Yes. Select a Grok video mode on this page and the request JSON, field list, and pricing panel will update to match that model.

Start building

Turn Grok video into your production workflow

Open the generator, test a Grok video prompt or source image, then use the API docs when you are ready to wire it into production.

API-ready
workflow
4 modes
video tools
Task ID
follow-ups