2,000 16:9 GPT Image 2.5 medium images on Sume: $120.00
One 16:9 GPT Image 2.5 image at medium quality reserves $0.06 on Sume, so 2,000 cost $120.00. Per-quality price list, read 2026-10-09.
2,000 16:9 images from GPT Image 2.5 at medium quality cost $120.00 on Sume, at $0.06 per image. The same size at max quality costs $0.89 per image, 14.8 times as much.
Price per image by quality
These are the Sume reservations for one 16:9 image per job, with no input images, read 2026-10-09. Sume bills the provider's token-based list price times 1.25 and rounds up to a cent, so small prices step in whole cents.
| Quality | Per image | For 2,000 images |
|---|---|---|
| low | $0.03 | $60.00 |
| medium | $0.06 | $120.00 |
| high | $0.23 | $460.00 |
| xhigh | $0.40 | $800.00 |
| max | $0.89 | $1780.00 |
Choosing a quality
Pick the lowest quality that survives your use. Previews and thumbnails are a low-quality job; finals are worth the step up. Rounding matters at the bottom: a 16:9 low image is $0.06, so a cent of rounding is a large share of the price.
Model ids and sizes are in the Image API docs.
Sources
Related posts
More in Pricing
- A 20,000-character voiceover is about 22 minutes: render bill $2.30
One Sume TTS job can hold 20,000 characters, $0.95. At 15 characters per second that is 22.2 minutes, so a Timeline render reserves 23 minutes: $2.30. The math.
- $25 avatar budget: 9 standard, 6 plus or 3 max 15-second clips on Sume
$25 buys 9 standard (15 s x $0.184 = $2.76), 6 plus ($3.675) or 3 max ($8.25) avatar clips on Sume. How the remainders are spent and the mixed-tier option.
- 25 music tracks of 2 min: Sume $3.13 vs ElevenLabs $7.50
Sume Music charges a flat $0.125 per generation; ElevenLabs lists music at $0.15 per minute. 25 tracks of 2 min: $3.13 vs $7.50. Read 2026-10-09.
- 25 text-heavy posters on Ideogram 4.5: $0.94, $1.88 or $6.88
Ideogram 4.5 costs $0.0375 low, $0.075 medium and $0.275 high per image on Sume. Twenty-five posters cost $0.94, $1.88 or $6.88, a 7.33x spread.
Written by Sume