Grok updates video generation model with new reference features

The video generation model known as Imagine Video 1.5 has received several new features shortly after its initial launch. The system now supports image and voice references, allowing users to guide the generated output more precisely.

New input methods

Image and voice references became available immediately for certain subscription tiers in the United States, specifically SuperGrok Heavy and SuperGrok Plus. These features are accessible through the web platform and the iOS application, with a broader rollout scheduled for all user tiers over the following days.

Generation capabilities

In addition to reference inputs, the update introduces native 1080p generation and text-to-video capabilities. Users can now generate video directly from a text prompt without needing a starting image. This feature combines the platform's image generation technology with image-to-video processing. Native high-definition resolution and text-to-video generation are now generally available across the web platform, iOS, and Android applications.