Note: Grok Imagine is xAI's state-of-the-art video generation and editing model, announced in January 2026. Learn more in the Grok Imagine API announcement and Imagine API docs.
We've added Grok Imagine Video as a video model for both generation and editing. Create videos from text or images—with optional synchronized audio—or refine existing clips with restyling, object control, and scene adjustments.
About Grok Imagine
Grok Imagine is xAI's flagship video-audio model, designed for end-to-end creative workflows. Our integration brings the following benefits:
- Text-to-Video & Image-to-Video: Generate videos from a simple prompt or animate a static image. Best-in-class instruction following for precise control.
- Native Audio: Optional synchronized audio generation—voice, music, or sound effects created alongside the video.
- Video Editing: Refine existing clips in the video editor. Restyle scenes, add or remove objects, swap props, and control motion with a prompt.
- Cinematic Motion: Realistic motion, object interactions, and visual continuity. Support for zoom, pan, dolly, tilt, and timelapse.
- Flexible Output: 1–15 second clips at 720p, with multiple aspect ratios.
What's New
- Grok Imagine Video: A new video model for generation and editing. Available in the video model selector and as an edit option in the video editor.
How It Works
Generation: Select Grok Imagine Video when creating videos. Use text prompts or start from an image. Optionally enable audio. Choose duration (1–15 seconds) and aspect ratio.
Editing: In the video editor, select Grok Imagine as the edit model to restyle scenes, add or remove objects, change colors, or apply new visual styles—all via prompt.
Learn more about Grok Imagine and xAI at x.ai.