Gemini 3.1 Flash Image: 512px drafts at $0.045, 2K finals at $0.101
Google's per-image ladder runs $0.045 at 512px to $0.151 at 4K. Price a draft-then-final workflow: eight drafts and one final cost under half of nine 2K images.

Google's Gemini API page lists Gemini 3.1 Flash Image at $0.045 for a 512px image, $0.067 at 1K, $0.101 at 2K and $0.151 at 4K. Eight 512px drafts plus one 2K final cost $0.461 at those rates, against $0.909 for nine 2K images, so drafting small and finishing large halves the spend when you only keep one in nine.
What is Google's price ladder?
The page lists per-image equivalents for the model, which is billed in tokens: $0.50 per million input tokens and $60.00 per million output tokens on the standard tier.
| Output size | Price per image | Relative to 512px |
|---|---|---|
| 512px | $0.045 | 1.0x |
| 1K | $0.067 | 1.5x |
| 2K | $0.101 | 2.2x |
| 4K | $0.151 | 3.4x |
How much does draft-then-final save?
Take a creative loop where you generate nine candidates and keep one. Generating all nine at 2K costs 9 x $0.101 = $0.909. Generating eight at 512px and re-rendering the winner at 2K costs 8 x $0.045 + $0.101 = $0.461. That is about 49 percent less.
Two caveats. A re-render is a new generation, so the final may not match the draft; if you need the same image at higher resolution, an upscale is a different tool. And a 512px draft hides detail you may reject at 2K, so drafts suit composition and colour choices more than fine text.
How do I request a size on Sume?
The images API takes a resolution tier of 512, 1K, 2K or 4K, and n from 1 to 10 images per request. Which tiers a given model supports is in its endpoint record, and the price you pay is that record's pricing line times n, with Sume's margin already included, per the image docs. Google's direct per-image prices above are not Sume prices.
Image billing is all-or-nothing, so a failed draft is not billed. Check the endpoint's supported parameters first, because a tier a model does not list will not be accepted.
curl "https://api.sume.com/v1/images/models/$MODEL_ID/endpoints" \
-H "Authorization: Bearer $SUME_API_KEY"
# read supported_parameters (resolution values) and pricing[].cost_usd
# draft cost = cost_usd * n at the smallest tier you will useWhat should I watch for?
Reserve is estimated at submit, so a large n at a high tier needs more balance up front; a 402 insufficient_credits means the reserve did not fit, not that the images were generated. Keep drafts at the smallest tier, finals at the tier you will publish, and total the two before you start.
Sources
Related posts
More in Pricing
- Gemini 3 Pro Image: $0.24 at 4K vs $0.134 at 1K/2K, what to budget
Google lists Gemini 3 Pro Image (Nano Banana Pro) at $0.134 for 1K or 2K and $0.24 for 4K. How to budget 100 images and read the Sume price line.
- Gemini image batch is half price: what that means for Sume users
Google lists Gemini 3.1 Flash Image batch at $0.25 input and $30 output per 1M tokens, half standard. Sume documents no batch tier; here is how to plan.
- Gemini Omni Flash 1.1 at $17.50 per 1M tokens: dollars per second
Google prices Omni Flash 1.1 video output in tokens, about $0.10 a second at 720p. What that means per clip, and how Sume's per-second rate is built.
- Gemini Omni Flash 1.1 clip cost matrix: 3 to 10 seconds, 360p to 4K
A lookup table for Gemini Omni Flash 1.1: list cost per clip for 3, 5, 8 and 10 seconds at 360p, 720p, 1080p and 4K, with Sume's list x 1.25 beside it.
Written by Sume