Kling 3 with audio costs 50% more on Sume: 10 s is $2.10 vs $1.40
Kling 3 lists $0.112 per second silent and $0.168 with audio. Sume bills x 1.25: 10 s is $1.40 muted, $2.10 with audio. 100 audio clips of 10 s: $210.
The short answer
On Sume, a 10-second Kling 3 clip costs $1.40 without audio and $2.10 with audio, at both 720p and 1080p. The audio option adds 50 percent: the provider list rises from $0.112 to $0.168 per second, and Sume bills the list x 1.25. For 100 ten-second clips, that is $140 muted or $210 with sound.
Kling 3 prices by length, resolution and audio
In the catalog, Kling 3 is priced per output second and is flat across 720p and 1080p, so the resolution rows match. The audio column is the same clip with audio on.
| Resolution | Length | Muted | With audio | Audio adds | 100 audio clips |
|---|---|---|---|---|---|
| 720p | 5 s | $0.70 | $1.05 | $0.35 | $105.00 |
| 720p | 10 s | $1.40 | $2.10 | $0.70 | $210.00 |
| 720p | 15 s | $2.10 | $3.15 | $1.05 | $315.00 |
| 1080p | 5 s | $0.70 | $1.05 | $0.35 | $105.00 |
| 1080p | 10 s | $1.40 | $2.10 | $0.70 | $210.00 |
| 1080p | 15 s | $2.10 | $3.15 | $1.05 | $315.00 |
Arithmetic
Muted: $0.112 x 10 s = $1.12; x 1.25 = $1.40. With audio: $0.168 x 10 s = $1.68; x 1.25 = $2.10. The extra for audio is therefore $0.056 per second at list and $0.07 per second billable, or $0.70 on a 10-second clip.
Compare with models that include sound at no separate charge. Gemini Omni Flash 1.1 always has native synced audio (the API rejects generate_audio: false), and its 10 s 720p clip is $1.25. So at 720p an audio-on Kling 3 clip ($2.10) costs more than a 10 s Omni clip ($1.25), while a muted Kling 3 clip ($1.40) costs 15 cents more. Omni stops at 10 seconds per clip, whereas Kling 3 goes to 15.
Batch totals
Because Kling 3 is flat across resolution, the choice that matters is audio and length. 100 clips of 5 s are $70 muted and $105 with audio. 100 clips of 15 s are $210 muted and $315 with audio. A mixed batch of 40 audio clips and 60 muted clips, all 10 s, is 40 x $2.10 + 60 x $1.40 = $84 + $84 = $168.
When audio on is worth paying for
If you will replace the sound with a voice-over and music track anyway, generate muted and save $0.70 on each 10-second clip. If you plan to ship the clip as it comes, native audio saves the cost of a separate TTS job and Timeline pass, though TTS is cheap: 450 characters is 3 cents. Compare the whole pipeline before deciding.
The Video Router docs describe the generate_audio behavior per model; read capabilities from GET /v1/videos/models before you rely on a setting. See Video generation and the Video Router page.
Sources
Related posts
More in Pricing
- Krea Pro $21: 83 Seedance 2.0 videos is $0.25 each. Sume's?
Krea's page says Pro's 20,000 units make 83 Seedance 2.0 videos, about $0.25 each on annual billing. How to get the same number from Sume before you commit.
- Luma Ray 3.2 HDR costs 2x SDR and EXR 3x: 5 s at 720p and 1080p
Luma lists a 5 s Ray 3.2 clip at $0.30 SDR, $0.60 HDR and $0.90 HDR plus EXR at 720p; $1.20, $2.40 and $3.60 at 1080p. Sume does not list Luma.
- Magic Hour credit packs: $12 for 5,000 or $149.99 for 80,000
Magic Hour packs run 0.24 to 0.19 cents a credit; the plans work out near 0.1 cent. Why packs cost double, and how a USD balance on Sume differs.
- MAI-Transcribe-2 prices end with the year: budget 500 hours a month
Both MAI-Transcribe-2 prices are limited-time through year end. 500 hours a month is $50 batch, $270 streaming, $300 on Sume. Microsoft states no 2027 price.
Written by Sume