Build a scene with text to video
Describe one clear action: a camera moving through a room, a product rotating in studio light, or a character crossing a landscape. Choose a horizontal, vertical or square format to suit where the clip will be used.
Turn a scene in your head—or a set of reference images—into a short video. Create with Grok Imagine 1.5, with 6–15 seconds to tell your story.
A purchased plan or credit pack and sufficient credits are required. Welcome credits alone do not unlock this model.
Inspiration
Follow a car through a city at dusk. Watch how the camera, reflections and movement work together, then write a shot of your own.
Use the camera movement and lighting as creative inspiration for your own shot. The model version and settings for this clip are unverified.

Start with text when you want to explore a new scene. Add images when you already have a subject, product or visual direction. Both workflows use the same generator above.
Describe one clear action: a camera moving through a room, a product rotating in studio light, or a character crossing a landscape. Choose a horizontal, vertical or square format to suit where the clip will be used.
Upload one image to animate a visual idea, or add up to seven references to guide the subject and scene. In the Image to video tab, write what should change or move. Framing follows the reference image rather than a separate ratio setting.
Select Text to video, or choose Image to video and upload your references. Add a prompt that describes the subject, action and camera movement.
Choose 6–15 seconds and 480p or 720p. For text prompts, set the aspect ratio too. Review the credit quote and make sure your account has purchased model access.
Submit the shot, follow its progress and play the finished clip. Download the result or adjust your prompt for another version. Each new generation has its own credit cost.
Choose a sharp subject image with the framing you want. If you add more images, explain which subject, setting or visual detail each one contributes. Consistent lighting and a simple action make the result easier to judge.
References guide a generation; they do not guarantee an exact copy or a frame-by-frame sequence. This page has no last-frame control. Inspect faces, fine details and product text before publishing.
The length and resolution determine the quote. These estimates use the same current rates and minimum charge as the generator. Text and reference-image modes use the same credit calculation.
| Output | 6 seconds | 10 seconds | 15 seconds |
|---|---|---|---|
| 480p | 10 credits | 16 credits | 23 credits |
| 720p | 18 credits | 29 credits | 44 credits |
Start with a short 480p draft to check the motion, then choose 720p for the next version if you need more detail. The quote above the Generate button is the amount to review before each submission.
View plans and credit packsWrite the subject, one main action, the camera move and the lighting. These original text prompt ideas load into the generator for editing; clicking one does not start a generation.
A lone cyclist crosses a rain-soaked city street at blue hour. The camera tracks smoothly alongside at street level. Warm shop windows reflect in the puddles. Keep the movement slow and the cyclist clearly visible throughout one continuous shot.
A matte ceramic coffee cup rests on a pale stone table beside a window. The camera slowly pushes closer as a thin curl of steam rises. Soft morning light brings out the ceramic texture. One continuous shot, calm movement, no text or logos.
A small red fox walks through tall grass at sunrise and pauses to look toward the camera. A gentle tracking shot follows at eye level. Dew catches the warm light, with the distant trees softly out of focus. Natural movement in one continuous shot.
Using your own image? Upload it first, then describe the motion and camera change instead of rewriting every visible detail. Change one instruction at a time when comparing results.
No. Generating with this model requires a purchased plan or credit pack and enough credits for the displayed quote. Welcome credits alone do not unlock Grok. You can explore the settings and prepare a prompt before purchasing.
Yes. Choose Image to video, upload one or more images and describe the movement you want. This page accepts up to seven reference images. It uses their framing automatically, and does not provide a separate last-frame input.
xAI describes native sound effects, ambience and dialogue for Grok Imagine Video 1.5. The generation route used on this page does not expose an audio switch or audio uploads, so this page does not promise a particular soundtrack or lip-sync result. Use a workflow with explicit audio controls when sound is essential.
This generator supports 6–15 seconds at 480p or 720p. It does not offer 1080p or 4K for Grok. Other services may expose different versions or settings; the controls on this page define what you can generate here.
This page starts from text or reference images. It does not accept source videos for editing or extension. Open our Video editing or Video extension tool and choose a model that supports that operation.
Check the live credit table and the quote above the Generate button. Longer clips and 720p output use more credits. A minimum charge may apply. Your plan or credit pack determines the price you pay for credits.
The generator saves a pending draft before sending you to sign-in or checkout, then restores it when you return through that flow. A changed price or unavailable model is shown before proceeding. Keep the return link and allow browser storage for recovery.
When a submitted generation is marked failed, the task system refunds the credits charged for that task. A processing task is still in progress. If a completed clip will not play, check the existing task or download before submitting another paid generation.
Write a prompt or add your references, choose your settings and review the credit quote. Your next clip starts in the generator above.