tech
Imagine Video 1.5 with References
Our best video model, now with text, image, and voice references — generating up to 1080p.
TL;DR
- Imagine Video 1.5 now supports image and voice references.
- Users can generate video from text prompts alone.
- Native 1080p resolution is now supported for text-to-video and image-to-video.
- Image references can lock specific elements like faces, products, or locations.
- Up to seven references can be used per generation.
- Features are rolling out on grok.com, iOS, Android, and the xAI API.
- The API is live with the grok-imagine-video-1.5 model, with voice reference support available on request.