Kling AI / Video model

Kling v3

Kuaishou's cinematic video model with native audio, 1080p output, and elements that keep up to four characters or objects consistent.

Meet Kling v3

Kling v3 is Kuaishou's cinematic video model. Instead of loose reference images it uses elements: a character or object you define with a front view and optional extra angles, which it then keeps consistent through the shot. You can use up to four.

It comes in Standard and Pro tiers, renders at 1080p, and generates native audio that you can switch off when you plan to add your own voiceover or music, which also lowers the price.

Available in Advibly

Settings and estimated costs from Advibly's live model configuration.

Inputs
Text, Start frame, End frame, Up to 4 elements
Durations
3s to 15s
Aspect ratios
16:9, 9:16, 1:1
Output options
1080p
Quality tiers
Standard, Pro
Audio
Generated with the video, can be switched off
Estimated cost
$0.13 to $0.26 / secondOne credit costs $1. The range spans the cheapest and the most expensive settings this model offers; the composer shows the exact amount before you generate.0.13 to 0.26 credits / second

When to use it

Choose Kling v3 when the same person or product has to appear across several shots.

  • Series of ads with one recurring presenter
  • Cinematic camera moves and dramatic lighting
  • Silent b-roll to cut under your own audio

Things to keep in mind

In image-to-video the aspect ratio comes from your start frame.

Turn audio off when you will replace it anyway. It costs less.

Get more from your prompt

  1. 1.Give each element a clean front view plus one or two other angles.
  2. 2.Refer to elements by name in the prompt so it knows who does what.
  3. 3.Describe lens and camera movement in film terms: dolly in, low angle, shallow focus.
  4. 4.Use Standard to find the shot and Pro to finish it.

Common questions

What is Kling v3 best used for?

Kling v3 is particularly useful for series of ads with one recurring presenter, cinematic camera moves and dramatic lighting, silent b-roll to cut under your own audio. The right choice still depends on your input assets, delivery format, turnaround and credit budget.

What inputs does Kling v3 support in Advibly?

Kling v3 works from text, start frame, end frame and up to 4 elements. Advibly only shows the controls a model actually supports once it is selected, so you are never offered a combination it cannot render.

How many credits does Kling v3 use?

The current estimate is 0.13 to 0.26 credits / second ($0.13 to $0.26 / second, at $1 per credit). The exact amount follows the options you pick, and Advibly shows it on the generate button before anything is charged.

Reviewed September 29, 2026 against Advibly's model configuration and Kling AI's published material.

Developer source

Create with Kling v3

Open a new video workspace with this model selected.

Create with Kling v3
Back to video models