Video
Grok Imagine 1.5
Clips with sound from a prompt, a photo or reference images.
- Length
- 1 to 15 s
- Sound
- In every clip
- Resolution
- 480p, 720p or 1080p
- References
- Up to 7 images and a voice clip
- Aspect ratios
- 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16
Grok Imagine
xAI's Grok Imagine models on a Sceneform plan, separate from the Grok app. Video 1.5 makes clips of up to 15 seconds with sound. Image 2.0 makes and edits stills.
6 s · 480p
“Vertical. A boxer with long braids and taped fists hits a heavy bag in a dark gym under one hard red light. Slow motion on each punch: sweat sprays off her, the bag jolts, chalk dust hangs in the red beam. Close and low, the camera circles her. Sound: heavy thuds on the bag, her sharp breaths, the chain creaking.”
6 s · 480p
“Vertical. A parkour runner in a black hoodie sprints across a concrete rooftop at sunset and leaps the gap to the next building, landing in a roll. The camera races alongside him and tilts down into the drop as he flies over it. Orange sky, the city far below. Sound: fast footsteps, rushing wind, a friend shouting 'Let's go!'”
8 s · 480p
“Night in an empty parking lot under orange streetlights, a crowd in streetwear packed in a circle. 0-4 s: a candy-red lowrider with gold wheels and no plates bounces its front end on its hydraulics, higher with every hop. 4-8 s: on the last hop it stands straight up on its rear bumper, sparks fly off the asphalt and the crowd surges forward with phones in the air. Low angle from the edge of the circle, a fast push in on the final hop. Hydraulic pumps whining, chrome rattling, the crowd roaring, one shout: 'Higher!'”
6 s · 480p
“Vertical. Night on an outdoor basketball court under floodlights, chain-link fence, friends packed on the sideline. A player in a white tank top drives in and throws down a two-handed dunk, hanging on the rim as his friends storm the court. Low angle under the hoop. Sound: sneakers squeaking, the rim clanging, everyone screaming.”
6 s · 480p
“Vertical handheld selfie video in a kitchen at night. A guy in his twenties with bleached hair bites a tiny red chili pepper, chews, freezes, his eyes start watering and he says: 'Oh no. Oh no no no.' Shaky phone camera at arm's length. Sound: his voice, a friend laughing off camera.”
Make a Grok Imagine videoPlans from $19 a month
One name, two models.Video 1.5 and Image 2.0.
Grok Imagine is the name for both. Pick the video model for clips with sound, and the image model for stills and edits.
Images
Grok Imagine 2.0
Stills from a prompt, or edits to up to five of your images.
- Resolution
- 1k or 2k
- Per run
- Up to 4 images
- Edits
- Up to 5 images at once
- Prompt
- Up to 8,000 characters
- Aspect ratios
- 2:1, 20:9, 19.5:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:19.5, 9:20, 1:2
Start from a still.Then make it move.
Make the frame with Grok Imagine 2.0, or upload a photo. Grok Imagine 1.5 starts the clip on it and adds the sound.
The still16:9 · 2k
“Cinematic wide film still at night. A young woman in a silver puffer jacket and black boots sits on the hood of an old dark green pickup truck parked on a frozen lake, holding an unlit red road flare. Green and violet northern lights ripple across the whole sky and reflect in the dark blue ice. Snowy pine trees on the far shore. Cold, vivid, music video look. No text anywhere.”
The clip6 s · 480p · sound on · keeps the still's shape
“She stands up on the hood and strikes the flare, red smoke and sparks pour out as she holds it high, and the northern lights surge brighter above her. The camera slowly circles the truck. Sound: the flare igniting with a hiss, ice creaking, wind.”
Every clip has sound.Write what you want to hear.
- Speech
- "We missed our stop. Again."
- Effects
- She bursts out laughing.
- Ambience
- The train rattling over the tracks, the door chime.
Grok Imagine 1.5 makes speech, effects and ambience with the picture. Put spoken lines in quotes. Add a voice clip and an image to steer who speaks.
Weighing an alternative?Seedance and Kling run here too.
Sceneform runs Grok Imagine next to Seedance 2.5 and Kling 3.0. Paste the same prompt into any of them and keep the take you like.
“Golden hour at a backyard pool party. A man in a gold tracksuit runs across the flat roof of the pool house and cannonballs into the pool, and the splash soaks the friends cheering at the edge. The camera tracks him from the pool deck, then holds on the wave of water hitting the crowd. Sound: the crowd counting 'Three, two, one!', a huge splash, screams and laughter.”
Grok Imagine 1.58 s · 480p
Seedance 2.58 s · 480p
Kling 3.08 s · 720p
Grok Imagine 1.5
Pick it for short clips with sound, from one second up.
- Length
- 1 to 15 s
- Sound
- Always on
- References
- 7 images and a voice clip
Seedance 2.5
Pick it for one take of up to 30 seconds.
- Length
- 4 to 30 s
- Sound
- On or off
- References
- 30 images, 10 clips, 10 sounds
Kling 3.0
Pick it for up to six shots in one clip, and end frames.
- Length
- 3 to 15 s
- Sound
- On or off
- References
- 3 elements
How to useGrok Imagine
- 1
Pick how to start
Write a prompt alone, upload a photo as the first frame, or add up to 7 reference images and one voice clip. Name each reference in the prompt with @.
- 2
Write the shot
Say where it happens, who does what, how the camera moves and what you hear. Here is the lowrider prompt, part by part.
- Scene
- Night in an empty parking lot under orange streetlights, a crowd in streetwear packed in a circle.
- Subject and action
- 0-4 s: a candy-red lowrider with gold wheels and no plates bounces its front end on its hydraulics, higher with every hop. 4-8 s: on the last hop it stands straight up on its rear bumper, sparks fly off the asphalt and the crowd surges forward with phones in the air.
- Camera
- Low angle from the edge of the circle, a fast push in on the final hop.
- Sound
- Hydraulic pumps whining, chrome rattling, the crowd roaring, one shout: 'Higher!'
8 s · 16:9 · 480p - 3
Pick the length and resolution
Any length from 1 to 15 seconds. 1080p works from a prompt alone; a photo or references give 480p or 720p. The sound comes with the clip.
Simple pricing that scales with you.
Pick the plan for how much you post. Every model on every plan. Unused credits roll over. Cancel anytime.
Creator
$19/month
Billed monthly
For solo creators posting every week
500 credits every month
- Up to 166 Grok Imagine 2.0 images
- Up to 125 Grok Imagine clips
- Up to 1.5 hours of AI voiceover
- Roll over up to 1,000 credits
Pro
Most popular$49/month
Billed monthly
For creators posting daily across channels
1,300 credits every month
- Up to 433 Grok Imagine 2.0 images
- Up to 325 Grok Imagine clips
- Up to 4 hours of AI voiceover
- Roll over up to 2,600 credits
Studio
$99/month
Billed monthly
For brands and production teams
2,700 credits every month
- Up to 900 Grok Imagine 2.0 images
- Up to 675 Grok Imagine clips
- Up to 9 hours of AI voiceover
- Roll over up to 5,400 credits
Scale
$199/month
Billed monthly
For agencies running always-on content
5,500 credits every month
- Up to 1,833 Grok Imagine 2.0 images
- Up to 1,375 Grok Imagine clips
- Up to 18 hours of AI voiceover
- Roll over up to 11,000 credits
Every plan includesEvery model, 90+ and countingReusable AI personasSceneform Studio editorCommercial usage rights
Want checkout, comparison tables, or a custom volume? See the full pricing page
Grok Imagine FAQ
Grok Imagine is xAI's set of video and image models, also found in the Grok app. Sceneform is a separate studio that runs them next to Seedance and Kling, on a Sceneform plan rather than a Grok subscription. Grok Imagine 1.5 makes video clips with sound from a prompt, a photo or reference images. Grok Imagine 2.0 makes and edits images.
Plans start at $19 a month and include Grok Imagine video and images.
A 6-second clip costs 23 credits at 480p, 40 at 720p and 71 at 1080p. A Grok Imagine 2.0 image costs 3 or 4 credits. The $19 Creator plan has 500 credits a month, up to 125 one-second clips at 480p. The exact cost shows before you generate.
Any whole number of seconds from 1 to 15 per clip. Sceneform sets no daily cap on how many clips you make.
Sceneform blocks nudity and sexual content. Action, horror, swimwear and explicit lyrics are allowed. The model can still decline a prompt on its own.
Yes. Grok Imagine 1.5 makes speech, effects and ambience with the picture in every clip. Put spoken lines in quotes. Add a voice clip and an image to steer who speaks.
Yes. Upload a photo as the first frame and describe what happens next. The clip keeps your photo's shape and comes out at 480p or 720p.
Grok Imagine 2.0 is the image model. It makes stills at 1k or 2k in 13 shapes from 2:1 to 1:2, up to four per run, and edits up to five of your images at once.
Yes. Grok Imagine Edit takes a clip you upload and changes what your prompt describes. It works on up to 8 seconds of the clip.
Yes. Every paid plan includes commercial use and exports without a watermark.
Write the first shot.
