Grok Imagine Video
Grok Imagine reference-to-video at 480p or 720p, 1-15s: guide the shot with up to three reference images. Output is billed per second of generated video and every reference per input image.
Examples
representative generations · starting points to seed the playground
Cinematic scene combining these references into one coherent shot. Cinematic subtle motion: gentle camera push-in, natural life in the scene (light shifts, soft breeze, fine movement), the subject stays coherent and photoreal. Elegant, filmic, high quality.
The woman sits on the back of the teal vintage convertible and waves her hand cheerfully, her smile widening; her patterned headscarf and the hem of her light-blue dress flutter in a warm breeze, palm fronds sway, and fluffy white clouds drift slowly across the bright blue sky over the mountain. Sunny retro California glamour, subtle slow camera push-in, glossy nostalgic color.
In the rural garden scene, a burlap-faced scarecrow in denim overalls sits on a stool reading a book beside an antique brass gramophone, overgrown green yard with a weathered barn behind. The vibrant orange-and-pink powder butterfly flies INTO this scene and flutters gracefully AROUND the seated scarecrow — circling, hovering and drifting near it — its wings flapping at a steady, even, consistent rate throughout. Crisp high-resolution, soft natural daylight, gentle cinematic motion. No text, no watermark.
Per-model pricing
Transparent, usage-based rates for every xAI model.
Grok Imagine Video 1.5
Video
Image to Video
Grok Imagine Video
Video
Text to Video
Image to Video
Reference to VideoCurrent
Video to Video
Grok Imagine Image Quality
Image
Text to Image
Image to Image
Grok Imagine
Image
Text to Image
Image to Image