> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Kling 3 Pro (Audio)

> Kuaishou's flagship Kling with native audio on Kyma. Same visuals as Kling 3 Pro plus synchronized sound, pay per second.

## Overview

`kling-3-pro-audio` is `kling-3-pro` with native audio. Same photoreal humans, same smooth motion, same sharp output — plus synchronized ambient sound and dialogue. About 50% more expensive per second than the silent variant.

## Specs

| Field             | Value                                                                          |
| ----------------- | ------------------------------------------------------------------------------ |
| Model ID          | `kling-3-pro-audio`                                                            |
| Creator           | Kuaishou                                                                       |
| Best for          | Cinematic clips needing diegetic sound, talking-head shots, ambient atmosphere |
| Default duration  | 5 seconds                                                                      |
| Max duration      | 10 seconds                                                                     |
| Input modalities  | Text, image (I2V)                                                              |
| Output modalities | Video, audio                                                                   |
| Resolution        | 1080p                                                                          |
| Audio             | Native (ambient + dialogue, baked into the video)                              |
| Pricing mode      | Per second                                                                     |

## Pricing

|                 |   Cost   |
| --------------- | :------: |
| Per second      | \$0.2268 |
| Default 5s clip | \$1.1340 |

## Use this when

* The clip needs diegetic sound — espresso machine hiss, footsteps, dialogue, ambient cafe murmur.
* You're producing talking-head shots or atmospheric scenes.
* You'd otherwise have to add Foley in post.

## Pick something else when

* The clip is silent or you'll add audio in post: use [`kling-3-pro`](/models/kling-3-pro) and save \~33% per second.
* You want fast-cut social with bundled audio: use [`seedance-2-pro`](/models/seedance-2-pro) or [`seedance-2-fast`](/models/seedance-2-fast).

## Example

```bash theme={null}
curl -X POST https://kymaapi.com/v1/videos/generations \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-3-pro-audio",
    "prompt": "a barista pulling an espresso shot, the machine hisses, ambient cafe murmur",
    "duration": 5
  }'
```

The endpoint is async — POST returns `202` with a `job_id`; poll [`GET /v1/jobs/{id}`](/api-reference/videos-generations#poll-for-the-result) until status is `succeeded`.

## See also

* [Video Generation](/models/video-generation) — full family overview
* [`kling-3-pro`](/models/kling-3-pro) — same visuals without audio
* [`POST /v1/videos/generations`](/api-reference/videos-generations) — endpoint reference