Kling VIDEO 3.0

Video modelCard summarized

Text- and image-to-video with native audio, multi-shot and up to 4K

● Re-checked 12 Oct 2026 · 54 facts · 1 correction · 7 unconfirmedSource: Kling VIDEO 3.0 Model User Guide (official guide; Kling publishes no model card)

Use it via API

$0.168 / sec

Code example Compare

Where to call it

Kling AI APIkling-v3

Supports

Image input

Full feature support

Good for

E-commerce ads where on-image text, signs and logos must stay legible
Narrative shorts with dialogue, cross-cutting and voice-over
Bilingual or multi-dialect scenes

Overview

Kuaishou · 01Kling

Kling 3.0

Text- and image-to-video with native audio, multi-shot and up to 4K

Typical price

$0.168 / sec

1080p, native audio

Max resolution: 4K
Max length: 15 s
Modes: text-to-video, image-to-video
Released: 6 Feb 2026

Kling's user guide presents VIDEO 3.0 as the upgrade of VIDEO 2.6, built on a unified training framework that combines native audio with element (character and object) consistency control. New in 3.0: multi-shot generation, a start frame plus element references, coreference of three or more characters, multilingual dialogue with dialects and accents, and flexible 3–15 s output. Kling publishes no benchmark numbers. user guide

kling-v3

Typical price: $0.168 / sec
Max resolution: 4K
Max length: 15 s
Aspect ratios: 16:9, 9:16, 1:1
Sound: Generated with the video
Released: 6 Feb 2026

text-to-video
image-to-video
start and end frames
multi-shot (1–6 shots)
element references