Hedra API pricing: estimate endpoint vs Sume dry_run
Hedra says you can price the exact request with an estimate endpoint. Sume's hosted MCP has dry_run and max_spend_usd gates that preview cost before a job.

Hedra's pricing post says you can price the exact request with its estimate endpoint before running it. Sume's equivalent is on the hosted MCP server: a dry_run=true call returns an admission and cost preview without submitting the job, and max_spend_usd caps a real one.
Hedra's figures are from its post (read 2026-09-30); Sume's are from MCP tools and gates.
What does Hedra publish for a 10-second clip?
Hedra priced one fixed job on its API: 10 seconds, 1080p, native audio on, from a start frame. Its post lists a range from $0.60 on LTX-2.3 Fast to $7.91 on Seedance 2.5, and says nine of the 11 models it compared meet that spec. It also notes that a per-second rate means little without the resolution and audio setting attached.
| Setting | Value in Hedra's post |
|---|---|
| Duration | 10 seconds |
| Resolution | 1080p |
| Audio | Native audio on |
| Cheapest quoted | LTX-2.3 Fast, $0.60 |
| Most expensive quoted | Seedance 2.5, $7.91 |
How do I preview a cost on Sume?
The MCP safety gates table lists dry_run=true as an optional admission and cost preview that does not submit the job. The same page says to prefer generation_admission_preview and, or, dry_run before expensive bursts, and that ordinary single creates do not need that step.
Writes and paid tools also require an idempotency_key, which the docs describe as a dedup key, not human approval.
What stops a request from costing more than I planned?
max_spend_usd is an optional gate that is enforced only when you provide it. If you leave it out, nothing caps that call, so set it on scripted bursts. Pair it with a dry run on the first request of a batch to see what admission thinks it will cost.
Where do per-model prices live on Sume?
The video model catalog returns pricing_skus, described as pricing information per SKU, next to supported_resolutions and supported_durations. Read those before choosing a model; the field list is in Video generation. For a worked walk-through, see AI video API cost calculator: pricing SKUs.
Are the two features the same?
Not exactly. Hedra describes an estimate endpoint on its API. Sume's dry_run is documented as an MCP gate, so this post does not claim an equivalent REST estimate route. If you call Sume over REST, price from the catalog SKUs instead.
Sources
Related posts
More in Pricing
- HeyGen API concurrency limit: 1.5x burst vs Sume queued jobs
HeyGen Enterprise burst adds up to 50 slots billed at 1.5x. Sume queues jobs above plan concurrency and returns 429 only when the queue is full.
- HeyGen voice clone API: 20+ minutes, paid slot, no Sume clone
HeyGen's professional voice clone needs 1-10 recordings totaling 20+ minutes and a paid slot. Sume's TTS tool selects voices and has no clone upload.
- Lyria 3 Clip Preview 30-second price vs Sume Music fixed price
Google lists Lyria 3 Clip Preview (30s) at a per-song price. Sume Music charges one fixed price per accepted generation, whatever length the prompt asks for.
- MiniMax H3 Max reference images: 2 free, then per image
MiniMax includes the first 2 images free on H3 Max (5 on H3) and charges per extra image. On Sume, check the catalog for accepted reference types.
Written by Sume