xAI: Grok Imagine Video 1.0 Generations API

Grok-Imagine-Video is xAI's flagship multimodal model that pioneered native audio-visual synchronization. It generates 720p HD video clips up to 15 seconds with context-aware sound effects and dialogue in a single pass. Powered by the Aurora engine and trained on the world-class Colossus cluster, it prioritizes rapid inference and creative freedom. It supports text-to-video, image-to-video (first frame) and reference-to-video with up to 7 reference images, while maintaining xAI's signature permissive content policy.