3-minute explainer video, all in: $3.25 of API calls on Sume
A 3-minute explainer from voiceover, music, 8 stills, 6 short clips and a Timeline render costs $3.25 on Sume, itemized line by line.
A 3-minute explainer made from a 2,700-character voiceover, one music track, eight stills, six 5-second clips and one Timeline render costs $3.24825 on Sume in API calls. The Timeline render itself is $0.30, and the 6 clips are the largest line at $1.875.
The line items
Each line uses a list price from the Sume catalog. TTS is $0.0475 per 1,000 characters, music is a flat $0.125, Nano Banana 2.1 at 1K is $0.10 per image, Wan 3.0 at 480p is $0.0625 per second, a Timeline render is $0.10 per output minute (ceil), and compose is $0.02 per job.
The sum is $0.12825 + $0.125 + $0.80 + $1.875 + $0.30 + $0.02 = $3.24825.
| Item | Math | Cost |
|---|---|---|
| TTS narration, 2,700 characters | 2.7 x $0.0475 | $0.12825 |
| Music, 1 generation | flat | $0.125 |
| 8 stills, Nano Banana 2.1 1K | 8 x $0.10 | $0.80 |
| 6 Wan 3.0 clips, 5 s, 480p | 30 s x $0.0625 | $1.875 |
| Timeline render, 3 minutes | 3 x $0.10 | $0.30 |
| Timeline compose, 1 job | flat | $0.02 |
| Total | $3.24825 |
What is cheap and what is not
The voice, music, render and compose together are $0.57. The pictures and clips are the other $2.675. If you replace the six Wan clips with six more stills, the clips line falls from $1.875 to 6 x $0.10 = $0.60, and the total to $1.97325.
Raising the clips to 720p doubles that line to $3.75, and the total to $5.12325. Resolution on motion is the most sensitive choice in the stack.
What the total leaves out
The figure covers one pass. Retakes multiply the lines you retry, so a plan that expects two takes on every clip and three on the music is 12 clips ($3.75) and 3 music generations (0.375). Captions, trims and filters are separate small jobs at $0.02 each in the catalog.
A 3-minute render is exactly 180 seconds, so it bills 3 minutes. A 181-second render would bill 4 minutes and reserve $0.40; trim a second to keep it at $0.30.
- Render: ceil(minutes) x $0.10
- Compose: $0.02 per job
- Trim, filter: $0.02 per job
A second scenario
Scale to a month of explainers. Ten explainers of this shape cost 10 x $3.24825 = $32.4825, about $32.48, before retakes. If each needs one extra set of six clips, add 10 x $1.875 = $18.75, and the month is $51.2325. The clips are the line to watch; the voice and render lines barely move.
Next step
Use the Timeline docs page for the slot format, and the Jobs and results page for polling. Assemble from Sume-hosted media only: the Timeline requires Sume-hosted source URLs.
Reproduce the numbers
Open https://api.sume.com/v1/catalog, which needs no key, and find the entry for the product. Each model_pricing row carries price, unit, pricing_basis and the estimated, minimum and maximum cents. Multiply the unit price by your quantity and round only at the end. The catalog also has applied_* fields that include an agent fee of 550 basis points; the figures in this post use the base prices.
Sources
Related posts
More in Pricing
- 3 music tracks of 0.5 min: Sume $0.38 vs ElevenLabs $0.22
Sume Music charges a flat $0.125 per generation; ElevenLabs lists music at $0.15 per minute. 3 tracks of 0.5 min: $0.38 vs $0.22. Read 2026-10-09.
- 30-clip storyboard agent: which Sume plan accepts all jobs at once?
If an agent submits 30 separate generation jobs, only Startup (48 accepted) and above take them all at once; Pro (24) needs two waves of 18 and 12.
- A $30 wallet and a 24-clip batch: where the 402 lands on Seedance 2
With $30.00 and 24 Seedance 2 clips at $3.024 each, submit 9 is accepted and 10 returns 402, with $2.784 free. The reserve math and a stop-on-402 loop.
- 30-second 1080p AI video: Luma $7.20, LTX $3.90, Sume from $5.63
What 30 seconds of 1080p costs: Luma as six 5-second clips ($7.20) or three 10-second ones ($10.80), LTX Fast $3.90, and Sume. Arithmetic shown.
Written by Sume