Images & Video · Vendor docs · source Oct 1, 2026
Product Photo to Vertical Video Ad with Grok Imagine
Turn one product photo into an 8-second vertical video ad
“cost about $1.35 when they ran it”
Per source From the tool's vendor, not independently tested.
Copy-paste prompt
Here is my product photo and a one-line description: [what it is and who it's for]. Write a creative brief for an 8-second vertical (9:16) video ad: the product name, three different scenes to place it in, one camera move, a tagline, and a voiceover under 20 words. The product must look exactly as it does in the photo in every scene: same shape, color, label and parts. Then tell me which scene would sell it best and why.
The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].
What they did
xAI's Grok API cookbook includes an ad generator that starts from a single product photo. Grok (grok-4.7) writes a creative brief with three scenes, a camera move, a tagline and a voiceover. Grok Imagine edits the photo into three 9:16 scenes while being asked to keep the product exactly as it is. Grok compares them and picks the one that sells the product best, then the winner is animated into an 8-second video with a voiceover mixed in.
How it went
xAI says a run makes three images and one video, takes a couple of minutes, and cost about $1.35 when they ran it.
We only report what the source says. Results are self-reported unless noted.
Worth knowing
It's a developer template (Node.js 22.13 or later, an xAI API key, ffmpeg for the voiceover mix), not a click-and-go app. xAI's sample product is a travel mug, not a seller's listing.