Nano Banana 2 on Sume vs Google direct: the markup by resolution

Sume bills the provider list price times 1.25. Google's own price for the same Nano Banana 2 model is not the same as the list Sume pays, so the gap you see against Google direct is not a clean 25 percent. It changes with resolution. This post works out the exact gap at each size, using Google's published prices (read 2026-10-03) and Sume's rate tables.

The gap by resolution

Sume's provider list for Nano Banana 2 is $0.06 at 0.5K, $0.08 at 1K, $0.12 at 2K and $0.16 at 4K, which Sume multiplies by 1.25. Google's standard prices are $0.045, $0.067, $0.101 and $0.151.

What the numbers say

The percentage gap is largest at the small sizes (67 percent at 0.5K and 49 percent at 1K) and smallest at 4K (32 percent). In dollars the difference ranges from $0.03 at 0.5K to $0.049 at 4K.

So the premium is not a flat 25 percent over Google, because Sume's list basis is not Google's own list. The honest framing: Sume is more expensive per Nano Banana 2 image than Google direct, by $0.03 to $0.05, and you are paying for one wallet across Sume's models, one API surface, and no separate Google account.

When it matters

At 1,000 images a month at 1K the gap is $33 ($100 against $67). At 100,000 images it is $3,300 a month, which is worth a conversation. Below a few thousand images a month the gap is smaller than the time cost of managing a second vendor.

The snippet computes the monthly gap for your volume at each size.

google = {'0.5K': 0.045, '1K': 0.067, '2K': 0.101, '4K': 0.151}
sume = {'0.5K': 0.075, '1K': 0.10, '2K': 0.15, '4K': 0.20}
monthly_images = 1000
for size in google:
    gap = (sume[size] - google[size]) * monthly_images
    print(f'{size}: gap ${gap:.2f} per {monthly_images} images')