Text to video
Turn your prompt into a 1–15 second clip with optional generated audio.
Create from a prompt
IMAGINE WITH xAI
Create the scene. Change what happens next.
Generate videos from text or images with Grok Imagine Video. Edit existing clips, extend a scene and create short videos with audio on Insvee.
FROM IDEA TO MOTION
Turn your prompt into a 1–15 second clip with optional generated audio.
Create from a promptChoose a starting image. Describe how the subject and camera should move.
Animate a starting imageGuide a scene with up to seven images, for clips up to 10 seconds at 720p.
Add image referencesRework a source clip up to 8.7 seconds, keeping its duration and aspect ratio.
Edit a videoContinue a 2–15 second MP4 with another 2–10 seconds of action.
Extend a videoContinue the gentle tracking shot as the macaw banks towards the sunlit mist. Keep its colors, the waterfall and the camera movement consistent.
A CREATIVE DIRECTION TO EXPLORE
A good opening does not have to be the end. Use Video Extend to describe the next action, or Video Edit to rethink the scene you already have.
Try Video ExtendFIND YOUR GROK
| Feature | Grok Imagine VideoOn this page | Grok Imagine Video 1.5 | Grok Imagine Video 1.5 Lite |
|---|---|---|---|
| Best starting point | Generate, edit and extend | References and frame direction | Text or one starting image |
| Generated clip length | 1–15s; references up to 10s | 1–15 seconds | 1–15 seconds |
| Highest output | 720p | Native 1080p for text / start image | 1080p upscaled from 720p |
| Image guidance | Up to 7 references | Up to 14 references at 720p | One starting image |
| First / last frame | Starting frame | Start; ending frame / pair up to 720p | Starting frame |
| Source-video tools | Video Edit + Video Extend | — | — |
YOUR IDEAS. ROOM TO CREATE.
Compare credits, storage and capacity. Choose your Insvee plan.
Loading current plans and regional pricing…
A FEW USEFUL DETAILS
Yes. On Insvee, the classic Grok Imagine Video model has separate Video Edit and Video Extend tools. Editing accepts a source up to 8.7 seconds; extension accepts a 2–15 second MP4 and adds 2–10 seconds.
The classic model's frame workflow uses a starting image. Choose Grok Imagine Video 1.5 if you need a closing frame or a first-and-last-frame pair.
A prompt-only clip can be 1–15 seconds. Adding guidance images supports up to seven references and clips up to 10 seconds. Both offer output up to 720p. Create validates the selected combination before submission.
Open the model and tool in Create, choose your duration and output settings, and review the current credit estimate before generating. Applicable input-media charges are included. Editing and extension can include both source-video and generated-video costs. Insvee plans and credits are shown below.
The illustrative waterfall film was made with Wan 3.0, with the site owner's permission. It demonstrates a creative direction and is not a Grok output or a comparison of model quality.
CREATOR VOICES
Illustrative samples with fictional names. Not verified customer testimonials.
Essential cookies keep your account secure. Optional analytics and marketing cookies stay off until you choose.