Kling 4.0 brings native 30-second video generation
Kling has opened early access to Kling 4.0, which generates up to 30 seconds of video in one pass and accepts up to 15 reference assets, pushing its tool toward longer, more controllable footage.
Kling announced Kling 4.0, a major-version update to its video-generation model. The model is in early access, and the company says a wider rollout will follow in October. A faster, lower-cost variant, Kling 4.0 Flash, becomes available to Ultra Yearly subscribers on September 28.
The main change is native 30-second generation, so long takes and continuous scenes no longer have to be stitched together from shorter clips. Creators can set key moments with up to 10 keyframe images. Prompts can now run up to 8,000 tokens.
A new "omni reference" system accepts up to 15 assets in a single generation: up to 10 images, 5 videos (30 seconds combined) and 7 subjects, and a subject can be given a specific voice. Kling also describes targeted editing of expressions, movement, camera and backgrounds that leaves the rest of the scene stable. A "creative recreation" mode copies the pacing and shot composition of an existing video while swapping in different products or characters, and the company is pitching it at advertising.
Other additions include two-channel stereo audio, better lip-sync, a 21:9 aspect ratio, and broader language and dialect support covering Chinese regional dialects and several English accents. Some features are marked "coming soon": 10-bit HDR output at 4K and 1080p, and video extension up to 2 minutes.
The creation interface has been redesigned around a unified input box, a single editing timeline and a node-based canvas. The canvas includes an agent that interprets requests and uses tools to carry them out.
Kling did not publish benchmarks or pricing beyond the Flash subscription tier, and none of these claims have been independently evaluated.