Etsy-style listing gallery: 10 images per product, cost by image model
Ten gallery images for one product cost $0.07 to $1.00 on Sume depending on the model; 25 products is 250 images. Table with seedream, flux and gpt-image-2.5.

A ten-image gallery for one product costs $0.07375 on gpt-image-2.5 low at 1024 and $0.4375 on Seedream 5 Lite; a 25-product shop is $10.9375 on Seedream 5 Lite. The number depends far more on the model row than on how many angles you ask for.
Use the table to pick a lane per image type: hero, in-context, detail, flat lay. Hero shots can use a higher row while the filler angles use a cheap one.
Ten images per product
| Model row | Per image | 10 images | 25 products (250 images) |
|---|---|---|---|
| gpt-image-2.5 low, 1024 | $0.007375 | $0.07375 | $1.84375 |
| gpt-image-2.5 medium, 1024 | $0.0165 | $0.165 | $4.125 |
| flux-2-pro | $0.0375 | $0.375 | $9.375 |
| seedream-5-lite | $0.04375 | $0.4375 | $10.9375 |
| seedream-4.5 | $0.05 | $0.50 | $12.50 |
| gpt-image-2.5 high, 1024 | $0.065875 | $0.65875 | $16.46875 |
| nano-banana-2.1, 1K | $0.10 | $1.00 | $25.00 |
Keep the product identical across the set
Generate the hero once, then pass it back as a reference for every other angle. Seedream 5 Lite accepts up to 10 references and n up to 4, so one call can return four variants.
import os, requests
r = requests.post(
"https://api.sume.com/v1/images",
headers={"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"},
json={
"model": "bytedance-seed/seedream-5-lite",
"prompt": "Same ceramic planter, three-quarter angle on a walnut shelf",
"n": 4,
"input_references": [
{"type": "image_url", "image_url": {"url": "https://example.com/hero.png"}}
],
},
timeout=60,
)
print(r.status_code)
body = r.json()
if r.status_code == 200:
for item in body["data"]:
print(item["url"])
else:
print(body)Disclosure and file rules
Marketplace rules on sizes and AI disclosure change; Sume's own notes are in Etsy designed-by-AI disclosure if you want the record-keeping side. Check the marketplace page before upload.
A suggested split
Not every slot needs the same row.
- Hero (1 image): $0.0659 on gpt-image-2.5 high, or $0.10 on Nano Banana 2.1.
- Angles and context (6 images): $0.2625 on Seedream 5 Lite.
- Detail and flat lay (3 images): $0.0221 on gpt-image-2.5 low, before prompt tokens.
- That mix is $0.3505 per product.
Cost of the sample call
The n: 4 request above returns four images, so it costs four times the Seedream 5 Lite price: 4 x $0.04375 = $0.175. The gpt-image-2.5 rows are token-priced, so a prompt and any input reference add to the table figures; a 140-character prompt adds about $0.00025 per image.
Sources
Related posts
More in Use cases
- A first-frame still plus a 30-second Wan 3.0 clip: the pair's cost
Generate the opening frame with Higgsfield Soul, Seedream 5 Lite or Nano Banana 2.1, then animate it for 30 seconds on Wan 3.0. Totals at 480p to 1080p.
- Fix wrong text on a product label with ideogram-v4.5 and references
ideogram-v4.5 edits the first image and takes up to four more as references, five in total. Send a label photo plus a type sample; cost per edit by quality.
- Google Ads Shorts previews: render 9:16, 16:9 and 1:1 first
Shorts previews need a Video view or reach campaign with Shorts selected. Render all three Google sizes from one master with Sume Timeline for $0.30.
- Google Ads video rejects MP3, WAV and PCM: pair audio with a still
Google Ads video upload does not accept audio files such as MP3, WAV or PCM. Turn a radio spot into an MP4 with a still and your audio using Sume Timeline.
Written by Sume