Grok Imagine 1.5 prompt guide
Grok Imagine 1.5 from xAI makes stylised clips with sound from a prompt or a photo. It answers style words well, so say the look outright: anime, claymation, comic book or cinematic.
What Grok Imagine 1.5 takes
- Takes a prompt and a first frame.
- Clips of 5 or 10 seconds.
- Aspect ratios: 16:9, 9:16, 1:1.
- Settings: 480p; 720p; 1080p.
- From 5.75 credits per second.
The order to write a prompt in
- The subject and the one thing it does.
- The place and the light.
- The camera: the shot size and one move.
- The look: realistic, cinematic, anime, claymation and so on.
- The sound: what is heard, and any music.
Tips
- A clip of 5 or 10 seconds holds one or two actions: write one subject, one movement and one camera move.
- With a photo as the first frame, describe the motion, not the photo: the model already sees what is in it.
- Name the camera move, such as a slow push-in, an orbit or a static shot; otherwise the model picks one.
- Order events in time when there are two: first the door opens, then the dog runs out.
- Describe the sound as part of the scene, or ask for silence if you add your own audio.
- Draft with Grok Imagine 1.5 Lite at a much lower price a second, then make the final clip here.
Example prompts
A can of sparkling lemonade on a wet rock by a waterfall, drops running down the side as mist drifts past. Slow push-in. Bright daylight, crisp commercial look.
The curtains billow in the wind and sunlight sweeps across the room as clouds move outside. The camera holds still. Natural light, realistic motion.
A paper-cut fox runs through a forest of folded paper trees, leaves fluttering behind it. Side tracking shot. Handmade papercraft style, soft warm light.