Kling AI released new Video 2.6 model with native audio

Kling AI appears to be preparing a 2.6 release that bakes native audio directly into its video model. Internal references mention support for spoken dialogue, singing and sound effects under the tagline “See the Sound, Hear the Visual”, suggesting videos will ship with soundtracks generated in the same pass as the visuals, rather than relying on separate tools or manual editing.

UPDATE: Kling 2.6 was released

This release matters because it positions Kling 2.6 Pro competitively against other leading AI video models, including Sora 2 and Veo 3.1, by offering a full multimodal workflow (video + audio + image-to-video) with global-market audio support and high production quality.

The Kling family has moved from versions 1.6 through 2.5 across Turbo, Pro and Master tiers, steadily raising visual fidelity and control. For creators and studios using Kling 2.5 and Kling Omni through 3rd party platforms, where Kling 2.6 Pro is planned to be rolled out with built-in English and Chinese audio, this would close a long-standing gap versus rivals that already bundle speech and ambience. It would also build on earlier work which brought real-time audio generation to the broader Kling ecosystem.

Timing lines up with Kling Omni Launch Week, where the company, owned by Chinese short-video giant Kuaishou, is promising five days of new releases starting December 1, making a reveal around December 3 plausible within that window. With Kling O1 positioned as a multimodal hub for text, image and video creation, a 2.6 audio upgrade would fit as the sound layer of that stack, likely surfacing inside Kling’s web tools and partner integrations first, with wider availability depending on how the launch week unfolds.