Grok Imagine Video 1.5 takes up to seven references, in 1080p

xAI updated Grok Imagine Video 1.5 on 31 July, adding text-to-video and reference-to-video to the image-to-video it already did.

Seven references

One generation can now take up to seven references, and they can be voices as well as images — a face, a product, a location and the voice that should speak. In the API, reference-to-video can also use preset voices.

1080p

Text-to-video and image-to-video now render at native 1080p. xAI's release notes add a detail worth knowing when writing prompts: text-to-video runs as a text-to-image step followed by image-to-video.

Who gets it

In the Grok apps — grok.com/imagine, iOS and Android — text-to-video and 1080p are generally available, and references are rolling out, first to SuperGrok Heavy and Plus subscribers in the US. In the xAI API, text-to-video and image references are live now; voice references are available on request.