See More. Hear More. Control More.
More stable motion, higher-quality stereo audio, more precise lip synchronization, and enhanced visual output bring AI-generated video closer to a complete audiovisual experience.
Explore Kling 4.0 video generation, multimodal references, keyframes, audio-visual features, and editing capabilities.
Get Ready for Kling 4.0A major upgrade in visual realism, creative control, and storytelling.
More stable motion, higher-quality stereo audio, more precise lip synchronization, and enhanced visual output bring AI-generated video closer to a complete audiovisual experience.
Combine up to 15 multimodal references to guide characters, objects, environments, actions, and visual details with greater consistency and control.
Create cinematic sequences with up to 10 keyframes and native 30-second generation for stronger story continuity.
The capabilities shaping the next generation of AI video creation.
| Capability | Kling 4.0 |
|---|---|
| Text-to-Video | ✓ |
| Image-to-Video | ✓ |
| First & Last Frames | ✓ |
| Multi-keyframes | ✓ Up to 10 keyframe images as input |
| Omni Reference | ✓ |
| Video Duration | 3–30 seconds |
| Resolution | 720p / 1080p / 4K |
| Dynamic Range | 10-bit HDR (1080p, 4K) |
| Aspect Ratio | 21:9 / 16:9 / 1:1 / 9:16 / Auto |
| Audio Output | Two-channel stereo |
| Languages | Chinese, English, Japanese, Korean, Spanish, Portuguese, German, French, Indonesian, and more |
| Dialects and Accents | Chinese: Beijing accent, Taiwanese Mandarin accent, Northeastern accent, Sichuan dialect, Cantonese, and more. English: American, British, Indian, and more. |
| Text Generation | Chinese, English, Japanese, Korean, Spanish, Portuguese, German, French, Indonesian, and emojis |
| Text Input Length | Up to 8,000 tokens |
| Reference Images Input | Up to 10 images |
| Reference Video Input | Up to 5 videos, with a combined duration of up to 30 seconds |
| Reference Audio Input | Voice reference only |
| Reference Subject Input | Up to 7 elements in total, including up to 3 video-based elements |
| Reference Capability Notes | Up to 15 items across images, videos, voice references, and elements |
| Video Editing | Supports up to 5 input video clips, including 1 primary video. Each clip must be 3–30 seconds long, and the combined duration of all clips must not exceed 30 seconds. |
Explore cinematic video concepts.
| Feature | Kling 4.0 | Kling 3.0 |
|---|---|---|
| Native Duration | 30s | 15s |
| Multimodal Reference | Up to 15 | Available |
| Keyframes | Up to 10 | Limited |
| Audio | Stereo Audio | Native Audio |
| Lip Sync | Improved | Available |
| HDR | 10-bit HDR | — |
| Editing | Enhanced | Basic |
| Video Extension | ✓ | ✓ |
Everything you need to know about Kling 4.0 AI video generation, availability, features, and capabilities.
Kling 4.0 is the next-generation AI video generation model from Kling AI, designed to improve visual realism, creative control, and storytelling capabilities. It introduces enhanced audiovisual generation, multimodal references, keyframe-based storytelling, and advanced video editing features.
Kling 4.0 is scheduled to launch in October. Kling 4.0 Flash is already available for Ultra annual subscribers through early access.
Kling 4.0 has not yet been fully released. Early access to Kling 4.0 Flash is currently available for eligible Ultra annual subscribers, while the full Kling 4.0 release is expected in October.
Kling 4.0 introduces improvements across visual quality, motion stability, audio-visual synchronization, reference control, and storytelling. Key upgrades include stereo audio, more accurate lip sync, 4K resolution, 10-bit HDR output, multimodal references, keyframe control, and video editing capabilities.
Kling 4.0 supports up to 10 reference images. These references can help guide characters, objects, styles, and visual details to create more consistent AI-generated videos.
Yes. Kling 4.0 supports up to 5 reference videos, with a total reference video duration of up to 30 seconds. Video references can help guide motion, appearance, and scene consistency.
Yes. Kling 4.0 supports video editing with up to 5 video inputs, including one main video. Each video can be between 3 and 30 seconds, with a total input duration of up to 30 seconds.
Kling 4.0 supports native video generation up to 30 seconds, allowing creators to produce longer cinematic sequences with stronger narrative continuity.
Kling 4.0 brings advanced references, realistic audiovisual generation, and cinematic storytelling control into one creative workflow.
Explore Kling 4.0