Cheapest legal Wan 3.0 call: 2 s at 480p is $0.125, 15x less than 30 s
Wan 3.0 runs 2 to 30 s. The minimum call costs $0.125 at 480p, $0.25 at 720p and $0.50 at 1080p. A curl body, and why client-side length checks beat a 400.

The cheapest legal Wan 3.0 call is 2 s at 480p: 2 x $0.0625 = $0.125. That is 15 times less than a 30 s clip at the same resolution ($1.875), which makes the 2 s call the right size for a key check or a smoke test.
Minimum versus maximum
Wan 3.0 accepts 2 to 30 s at 480p, 720p and 1080p, with per-second rates of $0.0625, $0.125 and $0.25. The OpenAPI request schema only says duration is an integer of at least 1, so a 1 s request passes the schema and is rejected by the model's own range. Check the length in your own code first.
| Resolution | Rate per second | 2 s call | 30 s call | Ratio |
|---|---|---|---|---|
| 480p | $0.0625 | $0.125 | $1.875 | 15x |
| 720p | $0.125 | $0.25 | $3.75 | 15x |
| 1080p | $0.25 | $0.50 | $7.50 | 15x |
The smallest request
This is the body to use for a first call or a health check. The response is a 202 with id, polling_url, status and model.
curl -X POST https://api.sume.com/v1/videos \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Idempotency-Key: $(uuidgen)" \
-H "Content-Type: application/json" \
-d '{"model":"wan-3.0","prompt":"A red ball rolls across a wooden floor","duration":2,"resolution":"480p"}'Gotchas
Other models have different floors: Seedance 2.5 starts at 4 s, MiniMax H3 at 5 s, Gemini Omni Flash 1.1 at 3 s. Never hardcode 2; read supported_durations for the model you picked.
The request schema rejects unknown fields (additionalProperties is false), so do not add a seed or size to a test call: both return 400 unsupported_parameter on every model.
Sources
Related posts
More in Developers
- Check a GPT Image 2.5 size before you call it: a Python validator
Sume rejects GPT Image 2.5 sizes that break four rules: multiples of 16, edge 3840, ratio 3:1, 655,360 to 8,294,400 pixels. A Python check, run on 8 sizes.
- Check your image cost table against Sume /v1/images/models cost_usd
Read pricing lines from GET /v1/images/models/{id}/endpoints and compare cost_usd with your own table before a batch. Sume lines already include its margin.
- CI guard: fail the build when a new Sume pricing_skus key appears
A node:test file fetches GET /v1/videos/models and fails when a model uses a pricing_skus key your cost code cannot total. Wan, Omni and H3 rates in the table.
- Run an Agent Completion nightly from CI with curl and a spend cap
A 25-line bash job that starts a Sume Agent Completion with a cap and an idempotency key, polls the receipt, and fails the build unless the run completed.
Written by Sume