Nano Banana 2.1 still plus Grok Imagine 1.5 clip: cost per animation

One animated image on Sume, made from a 1K Nano Banana 2.1 still and a 6-second Grok Imagine 1.5 clip, costs $0.175: $0.10 for the still and $0.075 for the clip. The clip is billed at $0.0125 per output second at every resolution, so the still tier and the clip length are the only two numbers that move the total.

The figures below come from the Sume rate card code in the repository, read on 2026-10-11. They are Sume billed amounts (provider list times 1.25), not provider list prices.

What does each half of the pair cost?

Nano Banana 2.1 is billed per image by resolution tier. Grok Imagine Video 1.5 is billed per output second, with a duration of 1 to 15 seconds and a default of 6 seconds when you omit it. The clip model takes exactly one input image, so the still is not optional: a request without one is rejected.

What does one animation cost by still tier and clip length?

Add the still price to the clip price. Because the clip rate does not depend on resolution, a bigger still raises the total by a fixed step while a longer clip raises it by $0.0125 for every extra second.

What do 100 animations cost?

Multiply the cell by the count. A hundred 1K stills with 6-second clips come to $17.50, and a hundred 2K stills with 10-second clips come to $27.50. The still is the larger share only for short clips: at 6 seconds and 1K it is $0.10 of $0.175, and at 15 seconds it is $0.10 of $0.2875.

Read usage.cost on each finished job to confirm the billed amount against the table above. If the number differs from this page, trust the job.

How do you call the pair?

Grok Imagine 1.5 on Sume is image-in only, so it cannot take a reference video or edit an existing clip. If you need that, or a longer shot than 15 seconds, price a different video model instead. This post prices only the two models named; it does not compare image quality or motion.

Read usage.cost on each finished job to confirm the billed amount against the table above. If the number ever differs from this page, trust the job, because rates are read from the live catalog at submit time.

When is this the wrong pairing?

Grok Imagine 1.5 on Sume is image-in only, so it cannot take a reference video or edit an existing clip. If you need native audio or a longer shot, price a different video model instead. This post prices only the two models named; it does not compare image quality or motion.