AI video with sound: what audio adds per second on Sume

Native audio is a switch on some video models and always on in others. See the per-second price with and without it for Kling 3 Pro, Omni Flash and H3 Max.

5 min readSume
All posts

"AI video with sound" is a common search, and the answer depends on the model. Reported by Magic Hour (read 2026-10-07), MiniMax H3 generates native stereo audio. On Sume, whether you pay for sound separately depends on how the model's price is built, and the catalog's GET /v1/videos/models response shows it for each one.

Audio price per second, read 2026-10-07

On Kling 3 Pro the audio switch adds $0.07 per second, so a 10-second clip is $1.40 silent and $2.10 with sound.

Video price with and without audio; Sume catalog read 2026-10-07
Model on SumeAudio settingPrice per second
Kling 3 ProAudio off$0.14
Kling 3 ProAudio on$0.21
Omni FlashAlways onSingle price, no switch
MiniMax H3 MaxNative stereo audioPart of the per-second price; see the models list

Choosing without overpaying

  • If you will add a voice or a music bed yourself, generate silent and pay the lower price. A TTS line is a few cents and a music track is $0.125.
  • If you want ambience or effects that match the picture, such as footsteps or a crowd, generated sound is the only way to get them in sync.
  • A model with always-on audio cannot be bought silent. Mute it in the timeline or detach the track with Audio Detach at $0.01.
  • Speech produced by a video model is not lip-synced to a face you supplied. For a specific mouth saying specific words, use a lip-sync model.

Check the live list

Prices move when providers change them. Call GET /v1/videos/models for the capability and price of each model before you budget a batch, and treat the table above as a snapshot dated 2026-10-07.

Sources

Related posts

More in Pricing

All Pricing posts

Written by Sume