xAI
Grok Imagine 1.0 Video
Text-to-video and image-to-video in a single Grok Imagine 1.0 Video API call.
Grok Imagine Video 1.5 turns a single reference image into a realistic short clip with synchronized audio. xAI's image-to-video model delivers lifelike motion, strong prompt adherence, and consistent identity at 480p or 720p, 1 to 15 seconds.
optional · up to 4096 chars

required · 1 image · public URL only
default auto
default 480p
1 to 15 · default 8
on by default
Try one of these prompts
Real-world workflows and production use cases you can build and ship with this model.
Pass one reference image and an optional motion prompt. Grok Imagine Video 1.5 returns a 1 to 15 second clip with lifelike movement and synchronized audio — identity, face, and silhouette stay consistent from the first frame to the last.
Generate a clipTurn product shots, portraits, and posters into TikTok, Reels, and Shorts content. Grok Imagine Video 1.5 generates the motion and the audio together in one request, so there is no separate voice, music, or sound-effects pass to wire up.
Automate promo clips and campaign assets from a single source frame. Grok Imagine Video 1.5 keeps the subject on-model across shots and renders realistic lighting and camera moves, so batch output stays usable straight out of the call.
Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.
| Category | Unit | Price |
|---|---|---|
| Grok Imagine Video 1.5 | ||
| 480p | 1 second | $0.08 80 credits |
| 720p | 1 second | $0.138 138 credits |
Grok Imagine Video 1.5 animates from one reference image while keeping face, body, and silhouette consistent across the clip. Real skin texture and natural motion, no morphing or clones — built for believable live-action results.
Grok Imagine Video 1.5 generates video and audio together in a single workflow — dialogue, sound effects, ambient room tone, and music — so you skip the separate audio-generation and post-production steps entirely.
Grok Imagine Video 1.5 bills per second by resolution, with no subscription. The cost scales linearly with the duration parameter, and failed jobs refund automatically — you only pay for clips that succeed.
Both turn images into short clips on reAPI. Grok Imagine Video 1.5 focuses on image-to-video realism with native synchronized audio; Seedance 2.0 spans more input modes and a higher resolution tier. Here is how the two compare on publicly documented behavior.
Comparison reflects publicly documented behavior at the time of writing. Model behavior and pricing can change; check the pricing card above and the API docs for current values.
Sign up and grab a key from the dashboard. Free credits cover your first Grok Imagine Video 1.5 calls — no card required.
OpenPOST to /api/v1/videos/generations with model = grok-imagine-video-1.5-beta and one image_urls entry. Grok Imagine Video 1.5 returns a task ID immediately.
OpenGET /api/v1/tasks/:id until status is completed, then download the Grok Imagine Video 1.5 output clip and ship it.
OpenCommon questions about this model.
Explore more models in the same category.
xAI
Text-to-video and image-to-video in a single Grok Imagine 1.0 Video API call.
Kuaishou
Reference image plus reference video — controllable motion transfer in one Kling Motion Control API call.
—
One song plus reference images becomes a finished music video in a single API call.
Alibaba
Text, image, reference video and editing in one Wan 2.7 Video API call — per-second pricing, 720P or 1080P.
Try it in the playground or grab an API key to integrate now.