40 menu photos to 5-second Grok clips: $2.80 on Sume
Animating 40 dish photos for 5 seconds each on Sume's Grok Imagine row is 40 x 5 x $0.0125 = $2.50 unrounded, $2.80 once each job ceils to the cent.

Forty dish photos animated for 5 seconds each cost $2.50 at Sume's Grok rate of $0.0125 a second, or $2.80 as billed, because every job's price is ceiled to the cent: 5 seconds is $0.0625, which becomes $0.07, and 40 x $0.07 is $2.80.
Why the Grok row fits a menu
The Grok row is image-to-video only, 4 to 15 seconds, at 480p or 720p, and you should read the row's generate_audio flag in GET /v1/videos/models before you count on sound. A menu loop is a good fit because the photo is already the shot you want and the model only has to add steam, a pour, or a slow push-in. Plan to add your own audio, for example a short music bed in the timeline.
| Seconds per clip | Unrounded per clip | Billed per clip | 40 clips billed |
|---|---|---|---|
| 4 | $0.05 | $0.05 | $2.00 |
| 5 | $0.0625 | $0.07 | $2.80 |
| 6 | $0.075 | $0.08 | $3.20 |
| 8 | $0.10 | $0.10 | $4.00 |
Avoiding the rounding gap
Pick a duration that lands on a whole cent to avoid the rounding: 4 seconds ($0.05) and 8 seconds ($0.10) are exact, while 5 seconds loses 0.75 cents on every clip. Over 40 clips that is 30 cents, so it rarely matters, but it is why the unrounded and billed totals differ.
Running the batch
Each photo needs a public HTTPS URL and one submit. Send a distinct Idempotency-Key per dish, for example a hash of the file name, so a crashed loop can be restarted without paying twice for the dishes that already went through. Keep the whole batch under your wallet balance, since a submit returns 402 insufficient_credits when the balance is below the job's reserve.
- One
POST /v1/videosper photo withmodel: "grok-imagine-video-1.5"and the photo inframe_imageswithframe_type: "first_frame". - Prompt per dish: one move, one light cue, for example
slow push-in, steam rising, warm window light. - Poll or use a
callback_url; a failed job releases its reserved credits where applicable.
Sources
Related posts
More in Use cases
- AI video with sound vs voiceover plus lip sync: which Sume route
Use a video model's native audio for ambient sound, and TTS plus lip sync for a talking face. Sume's docs say video models do not lip-sync a later voiceover.
- Best AI video model for a consistent character: reference rows
For one character across clips, use a Sume row with reference images: Wan 3.0 (10), Omni (10), H3 and H3 Max (9). Kling 3 takes none. Read 2026-10-13.
- Best AI video model for lip-synced dialogue clips on Sume
For speech that matches the mouth, Sume offers Omni (native synced audio), H3 Max Lip Sync (audio 5-14.8 s, 10 MB max), or an avatar route. Read 2026-10-13.
- Best AI video model for product videos: pick the row by input
For a product video, pick the Sume row by the photos you have: first frame, end frame or reference images. Limits for seven rows, read 2026-10-13.
Written by Sume