Kling V3 I2V
Kling V3 image-to-video model supporting prompt-driven animation from a single reference image with 3-15 second output duration. The public API supports image, prompt, negative_prompt, sound, cfg_scale, mode, aspect_ratio, and duration.
Authorizations
API Key authentication. Format: Bearer YOUR_API_KEY.
Body
Kling V3 image-to-video request. image and prompt are required. Supported public fields are image, prompt, negative_prompt, sound, cfg_scale, mode, aspect_ratio, and duration.
Reference image URL or Base64 string.
1"https://example.com/keyframe.jpg"
Video generation prompt.
1 - 2500"The subject slowly turns toward the camera while wind moves through their hair"
Negative prompt describing undesired elements.
1 - 2500"blurry, shaky, low detail"
Audio generation switch.
on, off "off"
Prompt adherence strength between 0.0 and 1.0.
0 <= x <= 10.7
Generation mode.
std, pro "pro"
Video aspect ratio.
16:9, 9:16, 1:1 "16:9"
Video duration in seconds. Public API accepts integer values.
3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 5