Largest single AI video job each Sume model allows, by price
The most one Sume video job can reserve ranges from $1.13 on H3 to $42.65 on Seedance 2.5 at 30 seconds and 1080p. Eleven rows.

The most expensive single job on Sume is a 30-second 1080p seedance-2.5 clip at $42.65; the cheapest maximum is minimax-h3 at 15 seconds and 768p for $1.13. Sume reserves the price when you submit, so this is the amount a single request can hold from your balance.
How the ceiling is built
Take each row's longest duration and highest resolution from the Video Router docs, multiply by the billable per-second rate (list times 1.25), and round up to the cent. Seedance rows use token pricing at 16:9; Kling uses the audio-on rate; Recast and Genjutsu take the source length as duration, so the number is for the longest source the row accepts.
Ceiling by row
| Row | Longest at top tier | Billable total |
|---|---|---|
| seedance-2.5 | 30 s, 1080p | $42.65 |
| higgsfield-genjutsu | 30 s, 720p | $25.54 |
| h3-max-recast | 30 s, 1080p | $16.88 |
| seedance-2 | 15 s, 1080p | $12.76 |
| wan-3.0 | 30 s, 1080p | $7.50 |
| seedance-2-fast | 15 s, 720p | $4.54 |
| gemini-omni-flash-1.1 | 10 s, 4K | $3.75 |
| kling-3 | 15 s, audio on | $3.15 |
| minimax-h3-max | 15 s, 1080p | $3.00 |
| seedance-2-mini | 15 s, 720p | $2.84 |
| minimax-h3 | 15 s, 768p | $1.13 |
Why this matters
A batch queue should know the worst case before it submits. If you fan out ten 30-second Seedance 2.5 jobs at 1080p, the reservation is $426.50, not the clip price you saw in a 5-second test. Check the balance against the sum of the reserves, not the sum of the expected captures.
A failed job does not capture; see failed jobs. But a job that succeeds at the ceiling does, so a typo in duration (30 for 3) is the cheapest bug to prevent with a pre-flight check.
A pre-flight rule
Read GET /v1/videos/models, take supported_durations and the matching pricing_skus, compute the cost in your code, and compare it with a per-job cap you set. Reject requests above the cap before they reach Sume.
Sources
Related posts
More in Pricing
- Listing narration in 3 languages: 1,200 characters, $0.17 on Sume
A 1,200-character listing description spoken in three languages is 3,600 characters of Sume TTS at $0.0475 per 1,000, which is $0.17 per listing.
- Lyria 3.5 costs $0.08 a song at Google; Sume charges $0.125
Google lists Lyria 3.5 at $0.08 a song; Sume's Music Router bills a fixed $0.125 per accepted generation. The gap is $4.50 per 100 songs. What it covers.
- MAI-Transcribe-2-Streaming $0.54/hour vs Scribe v2 Realtime $0.39
MAI-Transcribe-2-Streaming is $0.54 per hour (intro, through 2026), Scribe v2 Realtime $0.39, AssemblyAI Realtime $0.45. Sume has no live STT.
- MAI-Voice-2.1 at $22 per million characters: a 60-second script
MAI-Voice-2.1 lists $22 per million characters. What a 60-second avatar script costs in speech, against Sume TTS 1.0, and why the avatar seconds cost more.
Written by Sume