Animate a still for 8 seconds: Grok Imagine $0.10 or Omni Flash $1.00

Eight seconds of image-to-video costs $0.10 on Grok Imagine Video 1.5 (grok-imagine-video-1.5) and $0.30, $1.00, $1.50 or $3.00 on Gemini Omni Flash 1.1 at 360p, 720p, 1080p or 4K on Sume. Grok is the cheapest animation; Omni gives you sound and higher resolutions.

Grok Imagine is image-to-video only on Sume and takes no audio. Omni makes sound and also accepts text-only requests.

Prices

Rates are provider list per second times 1.25, rounded up to whole cents. The Grok row has a single rate; Omni has four resolution steps.

What each one needs

Grok requires image_url or a first frame. It has no end frame, no references, no audio switch and no aspect_ratio field, so you set the shape through the image you send. The duration is 4 to 15 seconds.

Omni accepts a first frame and an end frame, or references. Duration is 3 to 10 seconds, ratios are 16:9 and 9:16, and synced audio is always produced. A single reference image with no frame is treated as reference-to-video, so use the first-frame field if you want the still to be the opening shot.

• Grok: 480p or 720p, silent, no aspect_ratio field.
• Omni: up to 4K, always with sound, 16:9 or 9:16 only.
• Neither accepts a seed.

Request for Grok

A first-frame job at 720p:

{
  "model": "grok-imagine-video-1.5",
  "prompt": "Gentle parallax, the lanterns sway",
  "duration": 8,
  "resolution": "720p",
  "frame_images": [{
    "type": "image_url",
    "image_url": {"url": "https://example.com/lanterns.png"},
    "frame_type": "first_frame"
  }]
}

Pick by the output you want

If you are producing a quiet loop for a web page or an e-commerce tile, eight seconds of Grok at $0.10 is hard to beat. If the clip needs ambient sound, a different crop or 1080p, Omni at 720p is $1.00 for the same eight seconds. Run a Grok pass first if the still already composes well, and escalate only if the motion is wrong.