GenerationText or image in, clip out
The model starts from a prompt or a still photo and generates the frames of a short clip, inferring how things in the scene should move.
AudioSound made in the same pass
Voices, music and effects are generated together with the picture, so lips and actions line up without a separate dubbing step.
EditingEdit and extend video
The API can add, remove or restyle objects in a clip, change lighting or weather, and continue a clip from its last frame to chain longer scenes.
DistributionGrok app, X and partners
Consumers use it in the Grok app; developers call the xAI API or partner platforms such as fal.ai, ComfyUI, InVideo, Flora and HeyGen.