Kling AI's Kling 4.0 adds native 30‑second video generation
Kling AI said its Kling 4.0 Flash model, in early access now, generates 30-second videos with up to 15 reference inputs and syncs lip movement in native audio.
Kling AI said it opened early access to Kling 4.0 Flash for Ultra Yearly subscribers, ahead of a full Kling 4.0 launch in October. The model generates native video running up to 30 seconds in a single pass.
https://x.com/Kling_ai/status/2104596718067257458
The model supports up to 10 keyframes for multi-scene control and up to 15 multimodal references per generation, Kling AI said. It also adds 4K resolution, 10-bit HDR output and more accurate lip sync for dialogue-driven scenes.
Video clips can also be extended after generation and edited for camera movement, style or background without a full re-render, the company said. Kling 4.0 also adds support for multiple languages, accents and dialects.
https://x.com/Kling_ai/status/2104586933473247280
Kling AI did not detail Kling 4.0 Flash's pricing or usage limits beyond the Ultra Yearly tier. No independent reviewer has yet tested the claims against rival video models from OpenAI, Google or ByteDance.