Get API Key:
Visit API Key Management Page to get your API Key
Add to request header:
Authorization: Bearer YOUR_API_KEYVideo generation model name
kling-o3-reference-to-video "kling-o3-reference-to-video"
Text prompt, max 2500 characters. Required when multi_shot=false (default), provided by multi_prompt for each shot when multi_shot=true
Reference syntax: You can reference elements, images, and videos in the prompt using <<<xxx>>> syntax, for example: <<<element_1>>> walking in the scene style of <<<video_1>>>
2500"Keep the same motion style, change to a snowy background"
Reference video URL
Video format requirements:
"https://example.com/reference.mp4"
Whether to keep original sound from reference video
Options:
true: Keep the original video sound (default)false: Do not keep the original video soundtrue
Reference image URL array (style/scene reference)
Constraint: Image count + element count <= 4 when video is provided
["https://example.com/style.jpg"]Video duration (seconds), range 3~10
Note: Shorter than the 15-second maximum for text-to-video/image-to-video, max supported duration is 10 seconds
3 <= x <= 105
Video aspect ratio
Options:
16:9: Landscape video9:16: Portrait video1:1: Square video16:9, 9:16, 1:1 "16:9"
Resolution tier
Options:
720p: Standard 720P1080p: High quality 1080P720p, 1080p "720p"
Advanced parameters
HTTPS callback URL for task completion
Callback Timing:
Security Restrictions:
2048 charactersCallback Mechanism:
10 seconds3 retries after failure (at 1/2/4 seconds after failure)"https://your-domain.com/webhooks/video-task-completed"
Video generation task created successfully
Task creation timestamp
1757169743
Task ID
"task-unified-1757169743-7cvnl5zw"
Actual model name used
"kling-o3-reference-to-video"
Task type
video.generation.task Task progress percentage (0-100)
0 <= x <= 1000
Task status
pending, processing, completed, failed "pending"
Video task details
Task output type
text, image, audio, video "video"
Usage and billing information