xAI· Video modelNewGrok Imagine 1.5 is xAI's video model, and every clip comes with sound: dialogue, effects and ambience made with the picture. Write a prompt or add a photo, pick 3 to 15 seconds at up to 1080p, and see the credits first.
Made with these models
Original clips made with this model, each with its own sound. Read the prompt, or press Use this example to load the same settings.
Grok Imagine 1.5
The camera skims low over the waves beside the lead galley as dozens of oars dip in unison, throwing spray across the lens. A swell lifts the bow and slams it down. The camera rises as the fleet glides into the shadow of the stone arch and sunlight bursts through behind it. One continuous shot, heavy, believable momentum. Sound: pounding war drums keeping time, creaking oars, crashing surf, a captain shouting "Steady... ROW!"
Grok Imagine 1.5
She turns and walks toward the camera along the cliff edge with steady strides as a huge gust snaps the cape into a rippling crimson arc that fills the sky behind her. A wave bursts against the rocks and spray drifts up through the sunlight. The camera tracks slowly backward at a low angle. One continuous shot, realistic heavy-fabric physics. Sound: howling wind, silk cracking like a sail, booming surf. No music, no text.
Grok Imagine 1.5The camera skims low over the waves beside the lead galley as dozens of oars dip in unison, throwing spray across the lens. A swell lifts the bow and slams it down. The camera rises as the fleet glides into the shadow of the stone arch and sunlight bursts through behind it. One continuous shot, heavy, believable momentum. Sound: pounding war drums keeping time, creaking oars, crashing surf, a captain shouting "Steady... ROW!"
Grok Imagine 1.5She turns and walks toward the camera along the cliff edge with steady strides as a huge gust snaps the cape into a rippling crimson arc that fills the sky behind her. A wave bursts against the rocks and spray drifts up through the sunlight. The camera tracks slowly backward at a low angle. One continuous shot, realistic heavy-fabric physics. Sound: howling wind, silk cracking like a sail, booming surf. No music, no text.
xAI says version 1.5 improves motion, physics and audio over its previous video model.
Dialogue, effects and ambience come in the same pass as the picture and land on the action, xAI says, with clearer, better synced speech than before. Sound is always on and included in the credits.
Movement stays coherent across the clip, with fewer warps and more believable weight and momentum, according to xAI: useful for running, falling, pouring and other heavy motion.
Give it a start frame and the model keeps your picture's detail and lighting, so the clip continues the image instead of reinventing it, in the same shape.
Pick 3 to 15 seconds at 480p, 720p or 1080p, in five shapes for text to video. A 5 second 720p clip is 29 credits and takes about 1.5 minutes.
This page runs Grok Imagine 1.5, the one Grok model here. Use 480p for drafts, 720p for most posts and 1080p for the final cut; credits grow with length and resolution, up to 152 for 15 seconds at 1080p. For start and end frame transitions, use Veo 3.1 or Kling 3.0.
| Model | This pageGrok Imagine 1.5 |  Seedance 2.0 |  Kling 3.0 Pro |
|---|---|---|---|
| Credits per clip | 29 (5 s, 720p) | 62 (5 s, 720p) | 23 (5 s, 1080p) |
| Clip length | 3 to 15 s | 4 to 15 s | 3 to 15 s |
| Resolution | 480p, 720p, 1080p | 480p, 720p, 1080p | 1080p |
| Shapes (text to video) | 16:9, 9:16, 1:1, 4:3, 3:4 | 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 | 16:9, 9:16, 1:1 |
| End frame | No | Yes | Yes |
| Sound | Sound always on | Optional sound, no extra credits | Optional sound, extra credits |
| Typical wait | About 1.5 min per clip | About 4.5 min per clip | About 3 min per clip |
Credits per clip at the default length and resolution, without optional sound. Longer, sharper or with-sound clips cost more; the button shows the credits before you run.
Three steps from an idea to a clip with sound.
Describe the subject, motion, camera and sound. Add a photo as the start frame to animate it.
Choose length, resolution and shape, and make up to 4 clips per run; the credits on the button update as you go.
A Grok Imagine clip is ready in about 1.5 minutes. Play it with sound, download the MP4, or loop it with video to GIF.
Four tips, starting with xAI's own advice.
xAI suggests directing the shot in plain language: the camera move, the pacing and the sound design. For example: slow push in on a lighthouse keeper as the storm builds, waves boom against the rocks, he whispers, "Not tonight." A few clear words for each part work better than a long string of adjectives.
Speech is generated with the picture, so put the exact words in quotes and name the speaker: the street vendor grins and shouts, "Two for one!" A 5 second clip fits a line or two; give longer speeches 10 or 15 seconds. If a clip should have no speech, say so and describe the ambient sound instead.
With a start frame, the model already keeps your picture's detail and light, so spend the prompt on what changes: her hair lifts in the wind, she turns and laughs, the camera drifts left. Repeating what the photo shows adds nothing. Use a sharp picture in the shape you want; the clip keeps it.
xAI suggests chaining shots into longer scenes. Make each shot its own clip of up to 15 seconds, repeat the same style words and character description in every prompt, and start each one from a matching frame. Then join the clips in your editor. Make matching start frames with text to image.
Grok Imagine suits short clips that need sound and steady motion.
Vertical or square posts with built-in dialogue or effects.
Bring a portrait, a pet or a travel shot to life, keeping its look.
Product moments and spoken hooks in every common shape.
Chain short shots with a consistent look into a trailer or a story.
Credits, wait times, inputs, sound and commercial use.
Compare Grok Imagine with other top video models, or turn an image to video with any of them.
ByteDance · Video modelNewSeedance 2.0ByteDance's flagship: film-like camera work, big action and physics, sound at no extra cost. Does not take photos of real people.62 credits per 5 s clipKuaishou · Video modelNewKling 3.0Kling's top tier in 1080p: smooth, controlled motion; start and end frames.23 credits per 5 s clip
Google · Video modelNewVeo 3.1Google's full Veo: rich detail and natural light; start and end frames.33 credits per 4 s clip
MiniMax · Video modelHailuo 2.3Strong push-ins and lively faces; takes a few minutes.12 credits per 6 s clipWrite a prompt or add a photo and see the credits first.