Kuaishou's Kling AI has unveiled Kling 4.0 in preview, a flagship video generation model that represents one of the largest capability jumps in the lineup's history. The update spans clip length, reference handling, native resolution, and audio, addressing longstanding limitations that have hindered professional use of AI video tools.
The most immediate improvement is runtime. Kling 4.0 Preview extends native generation to up to 30 seconds per clip, with an experimental Long Video mode capable of producing continuous 120-second sequences at 1080p. Clips can also be extended to 60 seconds after generation. This directly responds to a frequent complaint about earlier Kling versions, whose shorter native runtimes made narrative and advertising work harder to plan around. For marketers and storytellers, longer native clips mean fewer awkward cuts and more coherent sequences, reducing the need for post-production stitching.
At the core of the update is Omni Reference, a system that allows a single generation to draw on up to 15 reference elements from a pool of as many as 50 uploaded files—30 images, 10 video clips, and 10 audio clips. Creators can lock a character's face across multiple shots, match a product's exact color from a reference photo, or carry a specific lighting setup from one scene to the next without re-describing those details in every new prompt. This level of control streamlines workflows and enhances consistency, a critical factor for brand storytelling where visual identity must remain intact.
Multi-shot generation builds on the same reference system: a single prompt can now produce a sequence of shots that hold spatial continuity—a room stays the same room as the camera moves, and characters keep the same clothing and face across cuts. That continuity has historically been a weak point for AI video tools, where scenes often visibly reset when the camera angle changes. By solving this, Kling 4.0 enables more professional-looking narratives and reduces the jarring inconsistencies that break immersion.
On audio, dialogue, ambient sound, and music are generated in the same pass as the picture rather than layered on afterward, with lip movement synced to spoken lines across multiple languages and accents. Different lines can be assigned to different characters within the same scene, removing a manual dubbing step that previously required separate audio software. This integrated approach saves time and ensures audio-visual harmony, making the output more suitable for global audiences and multilingual campaigns.
Native 4K output (3840×2160) rounds out the release, aimed at preserving fine texture, sharp edges, and depth of field through fast-motion shots rather than relying on post-generation upscaling. Creators evaluating AI video tools have increasingly flagged native resolution as the real test of whether a clip survives the move from preview window to actual publish. With true 4K, Kling 4.0 meets broadcast and high-end advertising standards, potentially opening doors to new professional applications.
For those looking to explore the model's capabilities, Kling 4.0 provides direct access to its generation modes as they roll out, including standard, Pro, and native 4K tiers. The site also offers a Kling 4.0 video prompt library for creators seeking practical starting points. By empowering creators, marketers, and small teams to test prompt-to-clip workflows, persistent-character generation, and multi-shot sequencing, Kling 4.0 Preview is poised to accelerate the adoption of AI video in professional content creation.
