Image upscale factor 4 on a 1024 square: about 16.8 MP for $0.20
Sume image upscale allows factor 1 to 4 at $0.20 per image, with about 16 megapixels reserved. A 1024 square times 4 is 4096 square, 16.78 MP. Check the math.
A 1024x1024 image upscaled by factor 4 is 4096x4096, which is 16,777,216 pixels or about 16.8 megapixels. Sume's Image Upscale 1.0 charges $0.20 per image and states that it reserves about 16 megapixels of generative output. The sizes line up, so a 1K input at factor 4 is the case the price was built around.
Input size versus output size
Arithmetic only: output side = input side x factor. The catalog does not publish a per-factor price or an output cap beyond the 16 MP reserve wording, so for larger inputs read the live catalog entry before running a batch.
| Input | Factor | Output | Megapixels |
|---|---|---|---|
| 1024 x 1024 | 2 | 2048 x 2048 | 4.19 |
| 1024 x 1024 | 4 | 4096 x 4096 | 16.78 |
| 2048 x 2048 | 2 | 4096 x 4096 | 16.78 |
| 2048 x 2048 | 4 | 8192 x 8192 | 67.11 |
Request fields
The strict body takes image_url (public HTTPS), upscale_factor (a number from 1 to 4), output_format (png, jpg or webp), metadata, and the communication fields. As with the other Sume model APIs, model and provider endpoint keys are rejected. Endpoint: POST /v1/image-upscale-1.0/upscale; poll the job and read the result.
curl -X POST https://api.sume.com/v1/image-upscale-1.0/upscale \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: upscale-hero-001" \
-d '{"image_url":"https://example.com/hero-1024.png","upscale_factor":4,"output_format":"png"}'Cost comparison worth knowing
Sume lists nano-banana-2.1 at $0.20 for a 4K image and $0.15 at 2K. So a 4K output costs the same $0.20 either by generating at 4K or by upscaling an existing image. If you already have the picture you want, the upscale keeps it; regenerating at 4K gives a new image.
Checklist
- Price is $0.20 per image, flat, in the catalog.
- Set
output_formatexplicitly if the next tool needs png. - Hosted MCP tool:
image_upscale_create, withidempotency_key.
Choosing a factor
A bigger factor is not always better. Factor 2 from a 2K image already gives 4096 pixels on a side. If the end use is a 3000 pixel print, an input of 1500 at factor 2 is enough, and you pay the same $0.20 either way because the price is per image. Because price does not rise with factor, the choice is about quality and file size: pick the smallest factor that reaches the target resolution and set output_format to jpg or webp when file weight matters. Every media job follows the same lifecycle: submit with an Idempotency-Key, receive a job, poll GET /v1/jobs/:id/status until it is ready, then read GET /v1/jobs/:id/result. A retry with the same key does not queue a second job, so a network error during submit never doubles a charge.
Sources
Related posts
More in Media tools
- Video inspect API: 24 stills max, so 2 fps covers only 12 seconds
Sume video inspect returns at most 24 stills per call and caps fps at 2. How to size frames.fps for a 60-second clip, plus fast versus precise seek.
- Is a 4:5 video a YouTube Short? 1080x1350 fits square or vertical
YouTube's Shorts page says square or vertical, up to three minutes. 4:5 is vertical, though not named. A 170-second 1080x1350 Timeline render costs $0.30.
- Keep 0.24 s of breath: shorten pauses instead of deleting them
Pad each kept range by 0.12 s per side and 9 pauses totalling 6.35 s lose 4.19 s, not 6.35 s. The speech keeps a 0.24 s breath at every join.
- Kling 3.0 Motion Control: a 13.2 s driver video bills 14 s, $2.205
Sume bills Kling 3.0 Motion Control at $0.1575 per output second, 1 to 30 s. A 13.2 s driver clip rounds up to 14 s, which is $2.205.
Written by Sume