5 channels, 60 thumbnails a week: monthly image bill on Sume

Five channels with 12 thumbnails each per week is 240 images a month: $36.00 on Nano Banana 2.1 at 2K or $13.35 on GPT Image 2.5 medium on Sume.

4 min readSume
All posts

The short answer

A team with five channels and 12 thumbnail candidates per channel each week generates 60 images a week, about 240 in a four-week month. On Sume that costs $36.00 per month on Nano Banana 2.1 at 2K (240 x $0.15) and $13.35 on GPT Image 2.5 medium at 2K (240 x $0.055625).

Thumbnail volume is predictable, which makes it easy to budget before the month starts.

The arithmetic

Twelve candidates per channel is three videos with four candidates each. The numbers scale linearly, so six candidates per video would double each total.

The table shows a four-week month. A calendar month of 4.33 weeks is about 8 percent more.

240 thumbnails per four-week month for five channels, as of 2026-10-08 (totals rounded to the cent; the Google row is a reference and not sold through Sume)
OptionCount x unit priceTotal
Nano Banana 2.1 at 2K240 x $0.15$36.00
Nano Banana 2.1 at 1K240 x $0.10$24.00
GPT Image 2.5 medium (2K)240 x $0.055625$13.35
GPT Image 2.5 low (1K)240 x $0.02475$5.94
Google direct, 2K standard (not on Sume)240 x $0.0504$12.10

How to run it

Give each channel its own prompt template and ask for n set to 4 per video. That is three calls per channel per week, and the cost is the same as 12 single calls.

Compare candidates at the size they will be shown at, since a thumbnail is judged small.

Retries and cost per channel

Some images will be rejected by eye, and rejected images are billed like any other, so budget a margin. The bullets add 20 percent to each Sume total in the table above (the totals multiplied by 1.2) and divide by the number of channels to give a per-channel cost.

Treat the per-channel figure as the number to compare across models, because it does not change when the count changes. Re-read the public catalog before a large run, since prices can change and the table is a snapshot from 2026-10-08.

  • Nano Banana 2.1 at 2K: $36.00 becomes $43.20 with 20 percent extra images, which is $8.64 per channel (a 5-channel job).
  • Nano Banana 2.1 at 1K: $24.00 becomes $28.80 with 20 percent extra images, which is $5.76 per channel (a 5-channel job).
  • GPT Image 2.5 medium (2K): $13.35 becomes $16.02 with 20 percent extra images, which is $3.20 per channel (a 5-channel job).
  • GPT Image 2.5 low (1K): $5.94 becomes $7.13 with 20 percent extra images, which is $1.43 per channel (a 5-channel job).

What the docs say to check

The n parameter takes up to 10 images per call, but per-model ceilings are lower, so read the n range descriptor for the model before you plan the number of calls.

The endpoint pricing lines are what Sume charges your wallet and already include the Sume margin, so a call costs cost_usd times n. Sume bills image generation all or nothing: a generation that fails or is cancelled is not charged.

Watch the wallet weekly rather than monthly. Sume reports the billed cost in each response, so a script can sum it per channel.

Every price in this post is a snapshot dated 2026-10-08. The Sume figures come from the public catalog and the Image API docs, the Google figures come from its Gemini API pricing page, and no figure here is a benchmark of quality or speed. Prices and model lists change, so confirm them on the day you spend money.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume