Models / Grok Imagine
How to prompt Grok Imagine
Last checked 2026-10-11Price checked 2026-10-11Grok Imagine Video 1.5 (xAI)
6 of 6 generation modes have a sourced guide. Modes the vendor docs don't confirm are shown greyed out, not guessed. Cheapest supported price: $0.02/s (Grok Imagine Video 1.5 Lite via xAI API).
Text to videotext-to-videot2v first generates a frame from the prompt then animates it: describe the opening frame clearly, then the motion, camera, pacing and sound.1 examples · checked 2026-10-11Image to videoimage-to-video (start frame)Send image (+ optional prompt) for image-to-video; output follows the image's aspect ratio (aspect_ratio ignored). 1.5 renders native 1080p. Addin1 examples · checked 2026-10-11Start & end framestart/end frameimage + last_frame pins both ends; prompt optional, steers motion/camera between them. Up to 4 keyframes at timestamp_s on a 1/3-s grid.1 examples · checked 2026-10-11Reference to videoreference-to-videoRefer to references by position: <IMAGE_0>, <IMAGE_1> ...; voices <AUDIO_0>. Up to 14 images and 3 preset voices on 1.5; up to 15 s,1 examples · checked 2026-10-11Video to videovideo-to-video/v1/videos/edits on grok-imagine-video: source .mp4 up to 8.7 s; output keeps source duration/aspect, capped at 720p. The model applies the change and1 examples · checked 2026-10-11Lip synclip sync / dialogue'The person from <IMAGE_0> ... speaking with the voice from <AUDIO_0>.' Audio is on by default.1 examples · checked 2026-10-11