Grok Imagine Video 1.5 Update: Leveling Up With 1080P, Voice Cloning, And More

Grok Imagine Video 1.5 Takes a Leap Forward

On July 31, 2026, xAI rolled out a game-changing update for Grok Imagine Video 1.5, adding four much-anticipated features since the model's debut in May. This major release expands on the platform with text-to-video generation, native 1080p output, voice cloning for scene consistency, and multi-reference scene control using up to seven visual anchors. For creative teams seeking a successor to Sora's consumer app, this update reshapes the landscape—offering new tools yet introducing fresh complexities.

The text-to-video generation and native 1080p output are now widely accessible on grok.com/imagine, as well as iOS and Android devices, opening up new possibilities for creators. However, more advanced features like voice cloning and comprehensive API access to image references remain exclusive to SuperGrok Heavy and SuperGrok Plus subscribers in the U.S., with a wider rollout anticipated soon. Developers interested in leveraging these tools must contact xAI's sales team for API-level voice reference access.

How the New Features Impact Professional Workflows

Before this enhancement, Grok Imagine Video was primarily an image-to-video platform, requiring a reference frame to kickstart the process. The new text-to-video approach removes this constraint, allowing users to create videos directly from text prompts. This method, akin to those offered by competitors like Google Veo 3.1 and Seedance 2.0, provides a more flexible video generation experience across supported platforms without extra costs.

"The addition of 1080p capabilities finally bridges the gap noted in earlier critiques," said a tech analyst. Although the technical details of how Aurora now supports 1080p remain undisclosed, the pricing structure offers insight. At $0.25 per second, the cost of 1080p is significantly higher than its 720p counterpart, reflecting the increased processing demands of this autoregressive model.

Voice Cloning: The Feature Everyone's Talking About

The voice reference feature is particularly noteworthy, promising consistency in both face and voice throughout generated scenes. Users can maintain character continuity, which is a boon for professional productions. This is achieved through a speaker embedding system that keeps the voice's unique traits intact, whether the character is delivering lines or simply interacting within a scene.

“Adding voice consistency to scene generation could revolutionize how we create content,” commented an insider from the creative industry.

However, access to this feature is currently limited, available only to select high-tier subscribers and requiring direct API requests for developers. Moreover, there are questions about consent and protection mechanisms, especially considering Grok's previous controversies involving non-consensual imagery generation.

Expanded Creative Possibilities with Multi-Reference

The multi-reference capability allows users to lock multiple visual components within a single generation, enhancing consistency across scenes. This could streamline workflows that previously needed separate tools, offering a more integrated solution for character continuity and scene setting.

Despite these advancements, free-tier users remain excluded from the new features as xAI continues to limit free access, aligning Grok Imagine Video with premium offerings in the market. The platform's competitive position remains strong, particularly with its new 1080p and voice capabilities, though affordability and ethical considerations linger as potential challenges.