Image-to-Video with xAI Grok Imagine Video v1.5

It's xAI's latest image-to-video model: upload a still image, describe the motion in a prompt, and it generates a video of up to 15 seconds at 480p or 720p, with synchronized audio included when generated. xAI describes the 1.5 models as its best image-to-video yet, with improved motion and physics. On Scenetra it runs as a node in the visual workflow editor.

On Scenetra: drop a Grok Imagine Video v1.5 node onto your board, upload the photo, write a short prompt describing the motion you want — camera moves, actions, atmosphere — pick a duration and resolution, and generate. No X app or API key needed; it works in the browser.

You can try it free online through Scenetra: new accounts get free welcome credits and a 7-day trial that covers your first generations. After that it's pay-per-generation — about $0.40 for a 5-second 480p clip — with no subscription gate and nothing to cancel.

$0.08 per second at 480p and $0.14 per second at 720p, plus a flat $0.01 per input image. A 5-second 720p clip is about $0.70 and a 10-second one about $1.40. Audio, when generated, is included in those rates.

Yes — v1.5 generates synchronized audio with the video, and it's included in the per-second price rather than billed as an add-on. That covers ambient sound and effects that match the animated scene.

V1.5 is image-to-video only, with better motion, physics, and audio, free-form 1-15 second durations, and a higher rate ($0.08-$0.14/s versus $0.05-$0.07/s). The original Grok Video also does text-to-video and is the cheaper option for drafts. Both are on Scenetra, so you can use the original for exploration and v1.5 for finals.

Describe the motion, not the image — the model already sees your upload. Name the action ('she turns toward the camera'), the camera move ('slow dolly in'), and the mood or sound you want. Short, concrete prompts tend to beat long scene descriptions, and Scenetra's board makes it cheap to iterate at 480p until the motion lands.