Kling 2.6 API: Native Audio, Specs, and Pricing

Kling 2.6 is Kuaishou's video generation model, released December 3, 2025. It is the first widely available text-to-video model with simultaneous audio generation: speech, dialogue, narration, sound effects, and ambience all rendered in a single pass with the visuals. This guide covers what Kling 2.6 has, how Standard and Pro modes differ, what it costs on Unifically and direct from Kuaishou, and how to call it.

TL;DR: Kling 2.6 generates 5- or 10-second MP4 clips at up to 1080p, 48 FPS, in 16:9, 9:16, or 1:1, with optional native audio (speech, SFX, ambience). Standard mode targets 720p; Pro targets 1080p. Unifically prices Kling 2.6 at $0.03 per second. Kuaishou direct lists Kling 2.6 Standard at $0.056/s without audio and Pro at $0.07/s without audio (or $0.14/s with audio).

What is Kling 2.6?

Kling 2.6 is the latest video generation model from Kuaishou's Kling team, released on December 3, 2025. It accepts a text prompt or a start image, an optional end frame, an optional audio toggle, and returns an MP4 with synchronized audio if requested.

The standout capability is single-pass audio-visual generation. Earlier video models either skipped audio or required a separate generate call. Kling 2.6 generates the visuals, voiceovers, sound effects, and ambient atmosphere together so they stay aligned to camera movement and on-screen action.

What's new in Kling 2.6

Kling 2.6 brought two practical capabilities at once:

Native audio in a single call, in both English and Chinese, covering speech, dialogue, narration, singing, rap, sound effects, and ambience.
An "Elements" feature for character consistency, combining up to four reference images so a recurring subject stays recognisable across separate generations.

Together those make Kling 2.6 a default for short-form content where audio cannot be a follow-up step (paid social, product demos, ad cutdowns). Kling also published the Motion Control variant for transferring motion from a reference video onto a character image. Useful for dance, walk cycles, and other choreography that is tedious to prompt from scratch.

What you can do with Kling 2.6

Two quality modes

Standard (720p). Lower-cost output for previews and high-volume social content.
Pro (1080p). Sharper detail tuned for paid placements and commercial use.

Both modes accept the same prompt surface. You pick the mode at request time.

Inputs

Text-to-video. Prompt is required; aspect ratio applies (16:9, 9:16, or 1:1).
Image-to-video. Supply a start image; prompt becomes optional steering text. Aspect ratio is implied by the start image.
Optional end frame. An end_image_url to suggest where the clip resolves.
Optional audio. Toggle audio generation on or off per call.

Output

Resolutions: 720p (Standard) or 1080p (Pro), with a maximum of 1080p.
Durations: 5 or 10 seconds.
Aspect ratios: 16:9, 9:16, or 1:1 (text-to-video).
Frame rate: up to 48 FPS.
Audio: native, single-pass; optional per call.

Kling 2.6 Motion Control. Transfer motion from a reference video onto a character image.
Kling 3.0. Newer Kling release with multi-shot timelines, 3 to 15 second single clips, and built-in sound generation. Sits above 2.6 in the lineup.
Kling 3.0 Omni. Image-to-video with synced audio, 5 to 10 seconds at 720p to 1080p.

Kling 2.6 pricing and how it compares

Unifically prices Kling 2.6 per second of generated video. Kuaishou's direct rates are also per second, so the comparison is apples-to-apples.

Source	Variant	Audio	Per-second price	5s clip	10s clip
Unifically	Kling 2.6	included or optional	$0.03/s	$0.15	$0.30
Kuaishou direct	Kling 2.6 Standard	without audio	$0.056/s	$0.28	$0.56
Kuaishou direct	Kling 2.6 Pro	without audio	$0.07/s	$0.35	$0.70
Kuaishou direct	Kling 2.6 Pro	with audio	$0.14/s	$0.70	$1.40

Source for Kuaishou direct rates: published Kling 2.6 API pricing on the official platform. Unifically pricing is the live kuaishou/kling-2.6 rate. Check the pricing page for the current value.

How to call Kling 2.6

The API is async: POST a generation, then poll a task endpoint until the MP4 is ready.

const API = 'https://api.unifically.com';
const headers = {
  Authorization: `Bearer ${process.env.UNIFICALLY_API_KEY}`,
  'Content-Type': 'application/json',
};

const start = await fetch(`${API}/kling-2.6/generate`, {
  method: 'POST',
  headers,
  body: JSON.stringify({
    prompt: 'Close-up of a barista pouring latte art, warm morning light streaming through the cafe window',
    duration: 5,
    aspect_ratio: '9:16',
    mode: 'pro',
    sound: true,
  }),
}).then((r) => r.json());

while (true) {
  await new Promise((r) => setTimeout(r, 3000));
  const task = await fetch(`${API}/v1/tasks/${start.task_id}`, { headers }).then((r) => r.json());
  if (task.status === 'completed') {
    console.log(task.video_url);
    break;
  }
  if (task.status === 'failed') throw new Error(task.error);
}

Things to know

Asking for 1:1 with image-to-video. The 1:1 aspect ratio applies to text-to-video. Image-to-video centres on the supplied frame, so the aspect ratio follows the image you uploaded.
Turning audio on when you don't need it. Audio toggling is per call. If your downstream pipeline replaces the audio anyway, leave it off. It can change pacing decisions the model makes.
Treating Kling 2.6 like Kling 3.0. Kling 3.0 adds multi-shot timelines and 3 to 15 second single clips. Kling 2.6 stays at 5 or 10 seconds with a smaller parameter surface. Don't migrate prompts blindly.
Stretching with multiple 5s clips when 10s is cheaper per second on competitors. Run the math against Veo 3.1 ($0.30/clip Fast, $0.60/clip Quality, 8s) before defaulting to Kling for longer narratives.
Forgetting Elements is a separate workflow. Character consistency via the Elements feature combines reference images and runs through its own playground tab, not a single text-to-video flag.

Frequently asked questions

What is Kling 2.6?

Kling 2.6 is Kuaishou's text-to-video and image-to-video model, released December 3, 2025. It generates 5- or 10-second MP4 clips at up to 1080p, in 16:9, 9:16, or 1:1, with optional single-pass audio generation.

How much does Kling 2.6 cost?

Unifically prices Kling 2.6 at $0.03 per second: $0.15 for a 5-second clip, $0.30 for a 10-second clip. Kuaishou direct lists Standard at $0.056/s without audio and Pro at $0.07/s without audio or $0.14/s with audio.

Does Kling 2.6 generate audio?

Yes. Kling 2.6 supports single-pass audio generation in English and Chinese, covering speech, dialogue, narration, singing, sound effects, and ambient soundscapes. Audio is optional and toggled per call.

What is the difference between Kling 2.6 Standard and Pro?

Standard targets 720p output for previews and high-volume social. Pro targets 1080p output for paid placements and commercial use. Both accept the same prompts. You choose the mode at request time.

How does Kling 2.6 compare to Kling 3.0?

Kling 3.0 adds multi-shot timelines, 3- to 15-second single clips, and built-in sound generation. Kling 2.6 sticks to 5 or 10 seconds with a simpler parameter surface and remains the lower-priced option for standard short-form work.

Kling 2.6 model page: live playground and full parameter reference.
Kling 2.6 Motion Control: transfer motion from a reference video.
Kling 3.0 and Kling 3.0 Omni: newer variants with multi-shot and sound.
Veo 3.1 vs SeeDance 2.0: head-to-head video API comparison.

Kling 2.6 API: Native Audio, Specs, and Pricing

What is Kling 2.6?

What's new in Kling 2.6