Grok Imagine 1.5 video generator

Grok Imagine 1.5 is xAI's AI video model. It makes 2–15-second videos at up to 1080p with sound, usually in one to three minutes, from text, a photo with keyframes, or up to 7 reference images.

Try it in your browser: new accounts get 40 free credits, enough for a 10-second 480p video. No card needed.

Try Grok Imagine freeCalculate the cost
Made with Grok Imagine 1.5 from one text prompt: 10 seconds at 1080p, with the sound generated with the picture. Turn the sound on to hear it.
Prompt
Cinematic close-up in a narrow Hong Kong alley on a rainy night, in the style of a 1990s Wong Kar-wai film. A woman in a dark red qipao stands under a dripping awning; red and green neon signs reflect on the wet walls and blur into soft bokeh. Rain streaks through the light. She slowly turns her head toward the camera, holds its gaze, and says quietly in English: "You came back too late." Slight slow motion, warm saturated colors, shallow depth of field, gentle handheld drift. Sound: rain on metal awnings, a distant tram bell, a faint old jazz record. No subtitles, no text.
Try this prompt

What can you make with Grok Imagine 1.5?

Single continuous shots with sound: character moments, comedy, fashion and mood pieces. Both clips below came from one prompt each, unedited.

A husky with stage presence

Dances to funk, freezes when the record scratches, then howls on cue. The howl is timed to its mouth.

Prompt
Vertical video, shot like a polished commercial. A fluffy grey-and-white Siberian husky with bright blue eyes stands up on its hind legs in a sunlit Scandinavian living room with pale wood floors, a big window and a record player. From 0 to 6 seconds a funky bass groove plays and, with a completely serious face, the husky dances to the beat: it wiggles its hips, bobs its head, waves its front paws side to side and does one wobbly spin, mouth closed. At about 6 seconds the needle scratches and the music stops dead. The husky freezes in silence for one second, mouth closed, staring straight into the camera with an offended look. Then it tilts its head back, opens its mouth and lets out one long, theatrical howl, "Awoooo!", like a singer hitting the high note. Lip sync: the mouth opens exactly when the howl starts, stays open and rounded while the howl sounds, and closes the moment the howl ends. No other sounds come from the dog. The dog keeps the same size, fur and markings the whole time, with natural paws. Locked-off camera at the dog's eye level, warm natural light, crisp detail. No text.
Try this prompt

Fashion film on the salt flats

Blue hour, a crimson silk dress lifted by the wind, sound generated with it. Vertical 9:16.

Prompt
Vertical fashion film at blue hour on vast white salt flats that mirror the pale pink and blue sky. A woman in a long flowing crimson silk dress walks slowly away from the camera toward the horizon; the wind lifts the long train of the dress into a soft ribbon behind her. The camera tracks smoothly behind her at a low angle. Minimal composition, the red dress as the only strong color, film grain, editorial look. Sound: steady wind, fabric rustling, footsteps on salt crust. No text.
Try this prompt

What is Grok Imagine 1.5 good at?

Speed, sound and control: it turns ideas around fast, speaks the lines you write, and follows a start frame, keyframes and an end frame.

Fast

A 10-second 1080p video usually arrives in one to three minutes, so trying another version of an idea doesn’t cost you the afternoon.

Sound and dialogue in the same pass

Speech, sound effects and ambience are generated with the picture; put spoken lines in quotes. Turn Generate Audio off for a silent clip at the same price.

Keyframes, not just a first frame

Give a start frame, up to 4 keyframes and an end frame. Keyframes are spread evenly through the video, so a 10-second clip takes all 4; they work best when they show the same scene.

Up to 7 reference images

Put characters, products and places in one video and mention them in the prompt as @Image1, @Image2 and so on.

Up to 1080p

480p and 720p for drafts and social posts, 1080p for the final cut. 1080p works from text or a single start frame.

2 to 15 seconds

Short clips and full 15-second shots, in Auto, 16:9, 9:16, 1:1, 3:2 or 2:3.

How much does Grok Imagine 1.5 cost?

4 credits per second at 480p, 8 at 720p and 12 at 1080p, with or without sound. 480p and 720p are open to free accounts; 1080p needs Basic or higher. Pick a length and resolution to see what it costs:

Resolution
This video costs
80 credits
40 free credits
Not enough: up to 5s at 720p
Basic, 600 credits a month
7 videos a month
Pro, 3,600 credits a month
45 videos a month
Ultra, 9,000 credits a month
112 videos a month

Rates as of October 30, 2026. Free credits refill to 40 every day. Free videos carry a watermark, are kept 7 days and are not for commercial use. See plans

In our AI video cost test (the same 6 prompts, 3 runs each, on 8 models at 720p), 97% of Grok Imagine 1.5's clips were usable, so a usable 5-second clip came to about 41 credits ($0.68), the 4th lowest of the 8.

Try Grok Imagine free

Grok Imagine 1.5 vs Seedance, Kling and Wan: which should I use?

Use Grok Imagine 1.5 for fast single shots with sound and keyframe control. Use Seedance 2.5 for multi-shot stories and the most references, Kling 4.0 for long takes with up to 8 keyframes, and Wan 3.0 for 30-second videos or videos made from a document. All of them are in the same generator and share one credit balance.

ModelBest forLengthUp toImages / videos / audio5-second video, credits
Grok Imagine 1.5Fast single shots with sound, keyframes, up to 7 images
Sound optional, same price
2–15s1080p7 / 0 / 020 at 480p
Seedance 2.5Cinematic quality, multi-shot stories, the most references
Up to 30 references, Auto ratio
4–30s1080p30 / 10 / 1060 at 480p
Kling 4.0Long takes with up to 8 keyframes
Sound always on
3–30s1080p10 / 5 / 365 at 720p
Wan 3.0Long single takes, videos from documents or web pages
Auto length, sound optional
2–30s1080p10 / 5 / 520 at 480p

Grok Imagine 1.5 specs

Made by
xAI
Version
Grok Imagine Video 1.5 (preview)
Modes
Text to Video, Image to Video (start frame, up to 4 keyframes, end frame), Multi Reference (images)
Length
2 to 15 seconds
Resolution
480p, 720p or 1080p (1080p from text or a single start frame)
Aspect ratio
Auto, 16:9, 9:16, 1:1, 3:2 or 2:3; Image to Video follows the start frame
Sound
On by default, can be turned off
References on VisionMorphic
Up to 7 images (JPG, PNG or WebP); no reference video or audio
Prompt length
Up to 4,096 characters

How do I use Grok Imagine 1.5?

Sign in, pick a mode, write a prompt and generate. Most videos are ready in one to three minutes.

  1. 1. Sign in with Google

    No card needed. New accounts get 40 free credits, enough for a 10-second 480p Grok Imagine video.

  2. 2. Pick a mode

    Text to Video takes a prompt. Image to Video takes a start photo, with keyframes and an end frame if you like. Multi Reference takes up to 7 images.

  3. 3. Generate

    The Generate button shows the credit cost before you submit. Failed generations return their credits.

Questions

Is Grok Imagine 1.5 free?

You can try it free at 480p and 720p. New accounts get 40 free credits and Grok Imagine uses 4 per second at 480p, so they make one 10-second 480p video, or 5 seconds at 720p. Free credits refill to 40 every day. Free videos carry a watermark and are not for commercial use; 1080p needs Basic or higher.

How much does a Grok Imagine 1.5 video cost?

4 credits per second at 480p, 8 at 720p and 12 at 1080p, with or without sound: a 10-second video is 40, 80 or 120 credits. Basic includes 600 credits a month, Pro 3,600. See all plans.

Can I animate my own photo?

Yes. Use it as the start frame in Image to Video, or as one of up to 7 reference images. People in your uploads must be you or have agreed to it.

Grok Imagine 1.5 or Seedance 2.5?

Grok Imagine 1.5 costs less per second and is strong at single continuous shots with sound. Seedance 2.5 handles multi-shot stories, long videos and many references better. Both are in the same generator and share one credit balance.

Is this the official Grok website?

No. VisionMorphic is an independent product. It is not affiliated with or endorsed by xAI.

Make your first Grok Imagine video

40 free credits make a 10-second 480p video with sound. No card needed.

Try Grok Imagine free