Kling 3 with sound costs 50% more: $0.14 vs $0.21 a second on Sume

kling-3 (Kling Video v3 Pro) on Sume lists at $0.112 per second with audio off and $0.168 with audio on, which Sume bills at $0.14 and $0.21 per second after the 1.25 multiplier. Sound adds exactly half again. For a 10-second clip that is $1.40 without and $2.10 with.

Clip prices (read 2026-10-04)

Each price is the provider list times 1.25, rounded up to the cent per job.

Set it explicitly

generate_audio defaults to the model's audio capability. If you add the sound later in your own edit, set it to false and avoid paying for audio you will replace.

curl -X POST "https://api.sume.com/v1/videos" \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-3",
    "prompt": "A violinist plays on a rooftop at dusk",
    "duration": 10,
    "resolution": "1080p",
    "aspect_ratio": "16:9",
    "generate_audio": false
  }'

Limits to remember

Kling 3 on Sume takes 4 to 15 seconds, 720p or 1080p, and 16:9, 9:16 or 1:1. It supports text, image and an end frame, and has no reference image or video inputs.

Kling 4.0

Kling 4.0 is not in the Sume catalog and the sources I read do not publish an API yet. Today's id is kling-3.