xAI's image- and text-to-video model, now with reference-based generation — pass in up to seven reference images to lock a face, product, outfit, location, or style across a freshly generated scene.
Last verified · 2026-08-01 · by Moe Ameen
Grok Imagine Video 1.5 is xAI's video generation model — the video half of Grok Imagine, the image-and-video creation feature inside Grok. It builds a clip from a text prompt or animates a still image into motion, with native synchronized audio (dialogue, sound effects, and music) produced in the same pass rather than added afterward. xAI shipped it as a preview in late May 2026 and moved it to general availability in mid-June 2026, where it topped the public Image-to-Video Arena leaderboard and undercut higher-end rivals sharply on price.
The notable addition — and the reason this page exists — is reference-based generation. You can pass in up to seven reference images, and each one locks a single thing in place: a person's face, a product, an outfit, a location, or a visual style. That lets you keep a character and swap the scene, keep the scene and swap the character, or hold both and change only the action. A voice reference can be supplied alongside a character image, so the same face and the same voice carry across every generated scene. It is the control layer generic text-to-video lacks: instead of hoping the model re-draws your subject the same way each run, you hand it the subject.
Treat the specifics as the current shape rather than fixed guarantees, since xAI iterates quickly. Generation runs at 480p and 720p at 24fps, with native 1080p available through the xAI API, short clip durations, and configurable landscape, square, or vertical aspect ratios (16:9, 9:16, 1:1). Access is through grok.com and the Grok apps — a free tier with a limited quota, and higher limits on the SuperGrok subscription — plus the xAI API, billed per second of output (roughly $0.08/sec at 480p and $0.14/sec at 720p at the time of writing). A speed-optimized "Fast" variant renders a short 720p clip in around 25 seconds. Confirm live specs and prices on xAI's own pages before quoting them.
Reference-based generation is what finally makes Grok Imagine Video 1.5 usable for a recurring brand: lock your product or your on-camera character across a batch of clips and the footage actually looks like the same thing every time. But a consistent clip is still just a clip — 720p, one aspect ratio, no captions, no hook frame, sitting in your downloads folder. Kompozy is the layer that turns that reference-locked footage into a published week of content across every feed.
Concretely, for a product or founder-led brand: generate three or four reference-consistent scenes in Grok (same product, different angles and settings), then drop them into Kompozy. Clipped Shorts cuts them into vertical 9:16 segments with word-synced branded captions for TikTok, Reels, and Shorts. Because the visual identity is already the real asset, Kompozy fans the same idea into a Carousel built with brand-exact HyperFrames, Quote Graphics of the key line, a Photo Post, a Blog Article, and an Email Newsletter — all governed by your Persona Brief so the copy stays on-voice while Grok's references keep the visuals on-brand. Autopilot and a per-post review pipeline then schedule and publish the set across the eight primary social platforms plus blog and email. Grok holds the character; Kompozy ships the campaign.
Grok Imagine Video 1.5 is xAI's video generation model, part of the Grok Imagine feature inside Grok. It generates a clip from a text prompt or animates a still image into motion, with synchronized native audio in the same pass. It launched as a preview in late May 2026 and reached general availability in mid-June 2026, topping the public Image-to-Video Arena leaderboard.
You can pass in up to seven reference images, and each locks one thing in place — a face, a product, an outfit, a location, or a visual style. That lets you keep a character and swap the scene, keep the scene and swap the character, or hold both and change only the action. A voice reference can travel alongside a character image so the same face and voice carry across scenes.
It generates at 480p and 720p at 24fps, with native 1080p available through the xAI API, and configurable landscape, square, or vertical aspect ratios. Audio is generated natively with the video — dialogue, sound effects, and music, including lip-synced speech. xAI iterates quickly, so confirm current limits on its own pages.
There is a free tier on grok.com and the Grok apps with a limited quota, with higher generation limits on the SuperGrok subscription. Developers can call it through the xAI API, billed per second of output. Check xAI for current tiers and pricing before budgeting.
Kompozy is the finishing and distribution layer. Generate reference-consistent clips in Grok, bring them into Kompozy, and it cuts vertical captioned shorts, reframes for each feed, spins the same idea into a carousel, quote graphics, a blog, and a newsletter in your Persona Brief voice, and publishes across nine destinations — eight social platforms plus blog and email.