Skip to main content

Kling v3

Kling v3 is Kuaishou’s Kling VIDEO 3.0 model. Use kling-v3 for text-to-video or image-to-video with first-frame or first-and-last-frame control, at 720p or 1080p, from 3 to 15 seconds, with optional native audio.
  • text-to-video: omit image_urls.
  • image-to-video: 1 image sets the first frame; 2 images set the first and last frame.
kling-v3 is a separate model key from kling-ai. Use Kling v3 Omni when you need multiple reference images.

Request parameters

Each request produces one video without a watermark. 4K, multi-shot storyboards, and element references are not offered.
Submit to POST /api/v1/task/submit/kling-v3. Save the returned data.task_id and poll /api/v1/task/status.

Pricing

Kling v3 is billed per second by resolution and audio.
Example: a 5-second silent 720p video costs 130 × 5 = 650 credits. /pricing remains the current source of truth.
See the Kling v3 API reference for the machine-readable Schema.