Start from text or a still
Pick Grok Video 1.5, then text to video or a starting frame according to what you have.
AI Video
Grok Video (Grok Imagine Video) is an AI video generator for text to video and image to video. The version here is Grok Video 1.5. Describe a shot, or upload a still and describe only the motion, to get a short clip back quickly enough to try several takes.
Pick Grok Video 1.5, then text to video or a starting frame according to what you have.
Name the subject, the camera and the light. If you uploaded a still, write only the motion.
Length and aspect ratio are chosen before the run. Keep the clip to one action.
Write the subject, the camera move and the one action in the clip. Grok Video 1.5 is fast enough that changing one phrase and generating again is cheaper than rewriting the whole description.
Upload a frame and describe the movement only: what enters, what drifts, where the camera goes. Describing the picture again tends to fight with the frame you provided.
One clear action holds up better than several at once. Slow continuous movement survives the render more reliably than fast cuts or crowds.
Grok Video is the Grok Imagine video model. Grok Video 1.5 turns a written description or a still into a short clip (text to video and image to video).
Yes. Upload a still and describe only the motion. Do not repeat the contents of the picture.
Keep it to one action. A longer duration stretches the same movement; it does not add plot.
Grok Image makes stills (text to image, image to image). Grok Video makes short clips. Generate a still on Grok Image if you need a first frame, then bring it here.