Cheapest possible clip per Sume video model: the minimum-length floor
The cheapest legal clip on Sume ranges from $0.11 (Omni, 3 s at 360p) to $1.07 (Seedance 2.5, 4 s at 480p). Per-second rates hide the minimum-duration floor.

Per-second rates make minimax-h3 and wan-3.0 look identical at 480p ($0.0625), but each model has a minimum duration, and that sets the cheapest clip you can buy. On Sume, gemini-omni-flash-1.1 at 360p for 3 seconds costs $0.1125, wan-3.0 at 480p for 2 seconds costs $0.125, minimax-h3 for 5 seconds costs $0.3125, seedance-2 for 4 seconds costs $0.70308 and seedance-2.5 for 4 seconds costs $1.074708.
The floor table
Each row is the lowest listed resolution at the shortest valid duration.
| Model | Lowest tier | Per second | Minimum seconds | Cheapest clip |
|---|---|---|---|---|
| gemini-omni-flash-1.1 | 360p | $0.037500 | 3 | $0.112500 |
| wan-3.0 | 480p | $0.062500 | 2 | $0.125000 |
| minimax-h3 | 480p | $0.062500 | 5 | $0.312500 |
| minimax-h3-max | 480p | $0.062500 | 5 | $0.312500 |
| seedance-2 | 480p | $0.175770 | 4 | $0.703080 |
| seedance-2.5 | 480p | $0.268677 | 4 | $1.074708 |
Why this matters for tests
If you only want to confirm that a prompt works, the minimum clip is what you pay for a smoke test. Smoke-testing seedance-2.5 costs about 9.6 times what an Omni smoke test costs ($1.0747 / $0.1125). Seedance 2.5 also bills a minimum of 4 seconds, so a 1-second request is not possible.
The floor is not the quality floor: 360p and 480p are preview-size frames.
Using it
Use the floor for integration tests of the job lifecycle, where the pixels do not matter. Use the full rate for finals. The list of valid durations per model comes from GET /v1/videos/models, described in the Video generation docs. One more rule: billing is linear in seconds inside each model's valid range, so a 6-second wan-3.0 clip at 480p is 6 x $0.0625 = $0.375.
Sources
Related posts
More in Comparisons
- Codex custom Responses provider: can it point at Sume?
Codex CLI 0.162.0 adds custom Responses-compatible provider config. Sume's Agent Completions is not a model endpoint; use Sume's hosted MCP from Codex instead.
- Context stitching and streaming TTS: Eleven v4 vs Sume async jobs
Eleven v4 lists context stitching and bidirectional streaming. Sume TTS is async and non-streaming, so keep a section in one job and join sections.
- Deepgram Flux Multilingual $0.0078/min: 7,500 minutes vs Sume STT $75
Deepgram lists Flux Multilingual streaming at $0.0078 a minute. For 7,500 minutes that is $58.50, against $75.00 on Sume STT. The full Deepgram ladder is below.
- ElevenLabs Scribe v2 $0.22/hour vs Sume STT $0.60/hour: 7.5 hours
Scribe v2 lists $0.22 per hour and Sume STT is $0.01 per minute, or $0.60 per hour. For 7.5 hours of audio: $1.65 vs $4.50, 2.7x apart.
Written by Sume