Kling AI releases version 4.0, adding stereo audio, multilingual lip sync, and a lighter Flash option.

Kling AI has released version 4.0 of its video generation model, adding longer single-pass clips, broader reference inputs, stereo audio and a lighter Flash variant. The update pushes the platform toward longer, more controllable output while the company positions it against professional production tools. Kling AI also frames the release as narrowing the gap between AI-generated and professional video, though its own comparison footage suggests the margin over competing Chinese models is thinning rather than widening.

Background: what the 4.0 release changes

The core generation limits have moved up. Users can now produce up to 30 seconds of video in a single pass, and the system accepts as many as ten keyframe images as input to lock down scene transitions, character states and story beats. Prompt comprehension has been expanded to 8,000 tokens, which matters for creators who write detailed shot descriptions rather than short prompts.

The upgraded "Omni Reference" system accepts up to 15 reference assets in one generation, according to Kling AI. That breaks down as up to 10 images, 5 videos with a combined maximum of 30 seconds, and 7 elements. Storyboards, wireframes, image grids, white-model videos and depth-map videos all work as references — a list that points at pipeline use, where a team feeds in pre-production material rather than describing a scene from scratch.

A "Creative Recreation" feature lets users take existing videos as templates and swap out products or characters. Kling AI says individual elements such as facial expressions, body movements, camera angles or backgrounds can be adjusted without affecting the rest of the clip. That is the kind of selective editing that has been difficult in generated video, where a change to one attribute often ripples through the whole frame.

How it compares with Wan 3.0 and Seedance 2.5

Comparison footage released with the update pits Kling 4.0 against Wan 3.0 and Seedance 2.5, both from Chinese competitors. Kling 4.0 holds its own but does not pull ahead, particularly on logical shot sequencing, physics and natural-sounding dialogue.

That result is worth reading carefully. The absence of a clear lead could be an early sign that the breakneck pace of progress in AI video models is starting to level off. The next Wan and Seedance releases will tell us more, and until those land the comparison is a single data point rather than a trend.

Audio, multilingual output and the features still pending

Kling 4.0 adds two-channel stereo audio and improved lip sync across more than ten languages, including German, English, Japanese, Korean and Spanish, along with various dialects and accents, according to the company. In-video text generation now works in multiple languages and stays stable during camera movements — a detail that matters for anyone generating signage, labels or on-screen graphics inside a shot.

Two headline capabilities are not yet available. 10-bit HDR output at 4K and 1080p, and a video extension function that would stretch clips to up to two minutes, are both listed as "coming soon" by Kling AI.

The Flash variant and developer access

Alongside the main model, Kling AI is shipping a lighter Flash variant. Kling 4.0 Flash is designed for faster generation at lower cost, though it tops out at 720p and 20 seconds of video. The trade-off is explicit: speed and price against resolution and duration.

The creation interface has been redesigned, and Kling 4.0 can now be accessed through MCP and CLI, which makes it possible to build automated workflows with AI agents. Pricing for the Kling 4.0 API has not been announced, with support listed as "coming soon."

What to watch next

The 30-second single-pass limit and 15-asset reference system are the numbers that will decide whether Kling 4.0 gets used for sequenced work rather than isolated clips. The pending items — 4K and 1080p HDR output, two-minute extension, and API pricing — are the ones that will determine whether it fits into existing production and automation pipelines. The competitive picture stays open until Wan and Seedance respond.