Price an AI video before you submit it on Sume: three ways
Read pricing_skus from the models endpoint, check the Videos panel Create estimate, or compute rate x seconds x 1.25. Worked examples.

You can price a Sume video before submitting it by reading pricing_skus from GET /v1/videos/models, by reading the credit estimate on the Create button in the Videos panel, or by multiplying the provider list rate by the seconds and by 1.25. All three should agree to the cent, because they use the same formula.
The video generation docs say the models endpoint returns capabilities and pricing, and that jobs reserve the amount at submit at provider list x 1.25, with usage.cost as the Sume billable amount on the result.
Worked example
Take a 7 second 1080p clip on Gemini Omni Flash 1.1. The fal list is $0.15 per second at 1080p, so list cost is 7 x $0.15 = $1.05, Sume cost is $1.05 x 1.25 = $1.3125, and the job reserves 132 credits after rounding up to the cent. The same formula for H3 Max at 768p, 7 s, is 7 x $0.08 x 1.25 = $0.70, or 70 credits.
| Request | List rate | Sume estimate | Credits |
|---|---|---|---|
| Omni Flash 1.1, 7 s, 1080p | $0.15 / s | $1.3125 | 132 |
| H3 Max, 7 s, 768p | $0.08 / s | $0.70 | 70 |
| Wan 3.0, 7 s, 480p | $0.05 / s | $0.4375 | 44 |
Where the estimate can differ
Two effects can make a quoted estimate differ from your mental arithmetic. First, rounding is per job, up to the next cent. Second, some modes add terms: a reference video on Seedance carries an assumed input length, and extra H3 references beyond the free ones add a fee.
Timeline renders have an explicit unbilled preflight, POST /v1/timeline-1.0/plan, per the Timeline docs, which compiles the document without charging.
Sources
Related posts
More in Developers
- Product hero video and LCP: poster, preload, no loading=lazy
A video can be the largest contentful paint. web.dev says to use a poster, preload it with fetchpriority high, and skip loading=lazy. Make the poster with Sume.
- Product photo variants in one call: the n parameter on Sume Images
Ask for several product photo variants in one POST /v1/images with n, read each model's ceiling from the catalog, and keep the cost predictable.
- Prompt cache diagnostics are GA: why did my video agent miss?
Anthropic took cache diagnostics out of beta on Sep 23 and OpenAI made them GA on Sep 8. What each reports, and where Sume's run receipt fits in.
- Prove a voiceover read the approved script: the transcript receipt
A source-bound Sume TTS job carries a transcript receipt with the revision, sentence IDs and a SHA-256. It proves the text sent, not how it was pronounced.
Written by Sume