Kling AI has released version 4.0 of its video generator, introducing higher resolution, precise controls, and a lighter Flash variant designed to close the gap between AI-generated and professional video. The updated platform allows users to generate up to 30 seconds of video in a single pass and accepts up to ten keyframe images to precisely manage scene transitions, character states, and story beats. Furthermore, prompt comprehension has been significantly expanded to 8,000 tokens to handle complex user instructions.
The upgraded Omni Reference system now accepts up to 15 reference assets in a single generation, including a mix of images, videos, and structural elements like depth maps and wireframes. A new Creative Recreation feature enables users to swap out products or characters in existing videos while selectively tweaking facial expressions, camera angles, or backgrounds without disrupting the rest of the clip. Alongside the main release, Kling AI introduced the Kling 4.0 Flash variant for faster, lower-cost generation, and added support for programmatic workflows through MCP and CLI integrations.
The update also brings two-channel stereo audio, improved multilingual lip sync across more than ten languages and dialects, and stable in-video text generation during camera movements. While 10-bit HDR output at 4K and extended two-minute video generation are listed as coming soon, the new features expand creative control and automation capabilities for video creators. This release matters because it equips professional and automated workflows with advanced multimodal controls, native audio, and developer tooling to streamline video production.