Skip to main content
Kling 3.0 is a single model version with multiple video capabilities. Through AnyFast, use model_name: "kling-3.0" with the corresponding Kling Legacy endpoint for each workflow.
These pages use Kling’s Legacy API fields such as model_name, prompt, image, and video_url. They do not use the upgraded API’s contents, settings, and options structure.

​
Supported capabilities

All three creation endpoints return an asynchronous task ID. Query the result through the matching endpoint action after submission.

Text to Video

Generate video from a prompt with multi-shot, camera, sound, and voice controls.

Image to Video

Animate a first frame, an end frame, or both, with optional motion-brush controls.

Motion Control

Transfer actions from a reference video to the character in a reference image.
  • Kling 3.0 Omni uses the separate kling-v3-omni model ID and /kling/v1/videos/omni-video endpoint for multimodal generation and video editing.
  • Element management creates reusable subjects that can be referenced by supported generation endpoints.
  • Voice management provides voice IDs that can be referenced in prompts when sound generation is enabled.

Kling 3.0 Omni

Generate or edit videos with text, images, videos, and reusable elements.