Grok Imagine Video 1.5 Explained: Text-to-Video, 1080p, and More

Grok's video generation tool just took a meaningful step forward. On August 1, xAI announced an expanded version of Imagine Video 1.5 — already its best video model — adding text-to-video generation, image and voice references, and native 1080p output. Here's everything you need to know about what changed and whether you can access it today.

▶ Watch Video on X

What exactly is Imagine Video 1.5, and what changed on August 1?

Imagine Video 1.5 originally launched last month as xAI's most capable video generation model, focused on more lifelike motion and improved audio quality. The August 1 update is a significant capability expansion on top of that foundation — adding three new features: text-to-video generation, image and voice references, and native 1080p resolution output. Previously, the model's maximum output was 720p.

What does text-to-video actually mean in practice?

Before this update, generating video in Grok required a starting image as input. Text-to-video removes that requirement — you can now describe a scene in a prompt and the model generates video directly from that text, no reference image needed. It's a workflow shift that lowers the barrier to entry considerably for users who don't have a suitable image on hand.

How do image and voice references work?

According to xAI, users can supply up to 7 reference images per generation to maintain consistent faces across different scenes. Voice references work similarly — you provide a voice sample and the model preserves that vocal identity across the generated video. This is particularly useful for anyone building multi-scene content where character or narrator consistency matters. Both features are live in the xAI API under the grok-imagine-video-1.5 model identifier.

Is 1080p a big deal compared to what was available before?

Yes, in practical terms. The previous maximum for Imagine Video 1.5 Preview was 720p, which is serviceable but limiting for anyone publishing to platforms where resolution is visible. Native 1080p output means the model generates at full HD rather than upscaling — a meaningful quality difference for motion-heavy content. For API users, 1080p output is priced at 8 credits per second, compared to 4.5 credits/second for 720p and 2.4 credits/second for 480p.

Who can use these new features right now?

As of August 1, the expanded capabilities are available to SuperGrok Plus and SuperGrok Heavy subscribers in the US on grok.com/imagine and iOS. SuperGrok is priced at $30/month; SuperGrok Heavy runs $300/month and supports up to 500 daily renders. Free accounts receive 5 generation credits per day but do not yet have access to the new features. A rollout to all subscription tiers is expected within the next few days across the Grok web app, iOS, Android, and the Imagine API.

Where exactly can you try it?

If you're on a qualifying tier, the features are accessible at grok.com/imagine and through the Grok iOS app. Developers can access text-to-video, image references, and native 1080p now via the xAI API using the grok-imagine-video-1.5 model. Android app support and broader tier availability are rolling out over the coming days.

The broader rollout timeline remains open — xAI said 'the next few days' without specifying a date. If you're on a lower tier and want access sooner, the API route is already live for developers regardless of subscription level.

This report was curated by the BASENOR Editorial Desk from the sources listed above. Read our editorial standards or email editorial@basenor.com to report an error.