Direct longer scenes with multimodal references, multiple keyframes, native stereo audio, and more stable motion. Kling 4.0 is in early access; generation is not yet available in Inspix.
30s
Native single-generation video
10
Keyframes for story control
15
Multimodal reference assets
Stereo
Native two-channel audio
Official Kling 4.0 launch video
Watch Kling AI’s introduction to the new model. Inspix generation remains unavailable until API access is ready.
Four announced upgrades
Kling 4.0 connects picture, sound, reference control, and narrative length. The result is a workflow built around a complete shot rather than a short visual fragment.
Audio + visual performance
Kling 4.0 is designed to keep fast action, continuous movement, tracking shots, and orbiting cameras stable. Its audio system also adds two-channel stereo and tighter lip-sync, so voices and performance can feel connected instead of layered together later.
Omni Reference
One generation can use up to 15 reference assets. Kling describes support for as many as 10 images, 5 reference videos with a combined duration of 30 seconds, and 7 defined subjects, including a voice attached to a subject.
Narrative control
Native 30-second generation gives a scene time for a setup, change, and payoff. Up to 10 keyframe images can identify the moments that matter, helping direct poses, expressions, transitions, and story beats across one continuous result.
Global creation
Kling 4.0 expands support for multiple spoken languages, regional accents, and dialects. The release also highlights more stable multilingual text, emoji, and logo rendering during movement and scene changes.
A practical workflow
These steps reflect the controls Kling has announced. Exact fields, limits, and billing will be confirmed when a public API is available for an Inspix integration.
Choose only the faces, products, environments, movement clips, and voices that must stay recognizable. Give every reference one clear responsibility.
Write the action in order, then place keyframes around the moments that cannot drift: an entrance, a reveal, a change of expression, or the final composition.
Inspect continuity, performance, dialogue, stereo sound, text, and product geometry across the complete clip before choosing what to edit or extend.
Production directions
The strongest use cases combine continuity, planned story beats, sound, and several visual anchors in one scene.
Use a longer native take and multiple keyframes to test blocking, performance, camera movement, and transitions before production.
Keep a product, presenter, environment, and campaign look connected through mixed image, video, and subject references.
Plan performance-driven scenes around native stereo sound, character voices, and closer lip-sync between dialogue and movement.
Develop versions for different markets with language, accent, dialect, text, and logo direction included in the creative brief.
Announced specification
The list distinguishes early-access capabilities from features Kling has explicitly marked as coming soon.
Kling 4.0 FAQ
Generate today
While the API integration is pending, you can use an available Inspix video model for current production work. This page will gain real controls only when generation can be completed here end to end.