Grok Imagine Video 1.5

What is Grok Imagine Video 1.5?#

Grok Imagine Video 1.5 is xAI's current Grok Imagine Video model. It replaces the original Grok Imagine Video generate endpoint with the same text-to-video and image-to-video flows, plus native synchronized audio, clips up to 30 seconds, and up to 7 reference images per call. Pass prompt alone for text-to-video, or add image_urls when you want a still to drive the opening frame or lock composition. Sound effects, ambient audio, music, and lip-synced dialogue generate in the same pass as the picture.

Key features of Grok Imagine Video 1.5#

Best for#

Use cases#

Run text-to-video when you only have a brief. A social team can prompt a 9:16 street scene with footsteps, breathing, and distant traffic in one call, no reference images required. For image-to-video, start with a clean still and let the prompt say what moves: a cosmetics brand can pass a packaging shot and get a presenter clip with products locked in frame, or a studio can animate a glass render so surfaces ripple and refract. With up to seven references, you can separate product, talent, and background while the prompt handles camera and audio. The 30-second ceiling and 24 fps make it a fit for ad spots, trailers, and social cuts rather than long-form video.

Limitations#

Negative prompts are ignored, so you describe what you want rather than what to avoid. Output tops out at 720p. Longer 30-second clips are more prone to artifacts than 5-to-8-second clips. Video extension still runs on the separate Grok Imagine Video Extend endpoint, not on this model.

Grok Imagine Video 1.5 vs Grok Imagine Video#

Grok Imagine Video 1.5 replaces the original Grok Imagine Video generate model. Both cover text-to-video and image-to-video, but 1.5 adds native synchronized audio, raises the per-clip duration from 10 to 30 seconds, and accepts up to 7 reference images instead of one optional still. The original model still handles Extend through its dedicated endpoint. Pick 1.5 when the clip needs sound baked in or a longer single pass; keep the original route only when you need Extend on an existing task_id.

API examples#

Call Grok Imagine Video 1.5 from any language by POSTing to /v1/tasks. Full parameter docs live at docs.unifically.com/models/video/xai/grok-imagine-1.5-video.

curl -X POST https://api.unifically.com/v1/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "xai/grok-imagine-1.5-video",
    "input": {
      "prompt": "A handheld tracking shot follows a skateboarder down a sunny street. He looks into the camera and says, Keep rolling. AUDIO: wheels on pavement, light wind, distant traffic.",
      "aspect_ratio": "16:9",
      "duration": 6,
      "resolution": "720p"
    }
  }'

Successful submission returns a task_id. Poll GET /v1/tasks/<task_id> or set a callback_url on the request to receive the finished video URL.