xAI Adds Image and Voice References to Imagine Video 1.5
The update adds text-only generation, native 1080p output and reference controls across consumer surfaces and the xAI API.
xAI announced on July 31 that Imagine Video 1.5 now accepts image and voice references, can generate video from a text prompt without a starting image, and supports native 1080p output. The release extends both the consumer Imagine experience and the company's video API.
What changed
The reference tools have a staged rollout. Image and voice references began in the United States for SuperGrok Heavy and SuperGrok Plus users on grok.com/imagine and iOS, with xAI saying other tiers would follow over the next few days. Text-to-video and native 1080p generation are already generally available on the Imagine website, iOS and Android.
For image-guided work, xAI says a generation can accept as many as seven references. Those inputs can preserve a face, product or location while the user changes another part of the scene. The voice workflow combines a character image with a voice reference so the same identity can carry across scenes.
Developers get a slightly different availability profile. Image references, text-to-video and native 1080p output are live in the xAI API under the `grok-imagine-video-1.5` model name. Voice-reference support in the API is available only on request.
Why it matters
Reference controls move video generation toward repeatable production rather than one-off prompting. Keeping a character, setting or product stable across shots is useful for advertising, storyboarding and longer sequences where visual continuity matters. The seven-image ceiling also gives creators a concrete boundary for planning multi-reference jobs.
The rollout still has meaningful limits. Consumer image and voice references began with two paid tiers in one country, while API voice references require separate access. The official announcement does not provide an independent quality comparison, latency measurement or broad availability timetable, so those questions remain open.
Status
Confirmed. xAI's official release establishes the shipped features and rollout boundaries. Internal confidence is medium because the available evidence is a single vendor source and the wider rollout has not been independently verified.
Sources
Update note: Last reviewed 2026-08-01. We will revise this post if xAI changes regional, tier or API availability.
Sources
- xAI — Imagine Video 1.5 with References — official
Drafted with AI assistance from source briefs; reviewed for citation completeness and label accuracy.