kling-v3-omni model. Combine a required prompt with first and last frames, reference images, or one reference video, then track the asynchronous result by img_uuid.kling-v3-omni model. Combine a required prompt with first and last frames, reference images, or one reference video, then track the asynchronous result by img_uuid.
@image_1 onward in first-frame, last-frame, then ref_imgs order. A reference video is @video_1, so stable array ordering keeps prompt references deterministic.
video_type: feature, the output follows reference motion and style; audio generation is rejected and multi-shot is forced on. With base, the source video is edited in place, first/last frames and multi-shot are unavailable, and existing audio is retained.
https://api.goenhance.ai/api/v1/videos/generations with required model: kling-v3-omni and a required prompt.image_url for a first frame. image_end_url requires that first frame; last-frame-only requests are not supported.ref_imgs values and refer to them with @image_N. The compatible alias image_urls is also accepted.ref_videos. video_url is a compatible alias.Authorization header.model to kling-v3-omni and write a prompt that explicitly addresses any @image_N or @video_1 material.feature or base mode./api/v1/videos/generations and persist the returned img_uuid.GET /api/v1/jobs/detail or provide a publicly accessible HTTPS custom_callback_url.Submit requests through GoEnhance at POST https://api.goenhance.ai/api/v1/videos/generations with model: kling-v3-omni. The response returns an img_uuid for tracking.
Pricing is per generated second. The exact tier depends on resolution, whether a reference video is used, and whether native audio is generated without a reference video. Multiply duration by the documented per-second rate.
Duration accepts integers from 3 through 15 seconds. Resolution accepts 720p, 1080p, or 4K.
The first frame is @image_1, the last frame follows when present, and ref_imgs continue in array order. Reordering the array changes the prompt references.
Not in video_type: feature, where audio generation is rejected. In base mode, the source video's existing audio is retained rather than generating a new track.
The requested 16:9, 9:16, or 1:1 ratio is used only when neither a first frame nor a reference video determines the output ratio.
No. Kling is associated with Kuaishou, but this page describes independent API access through GoEnhance. GoEnhance is not presented as Kuaishou or Kling's official API, partner, or model developer.
Poll when your worker already maintains a task loop. Use custom_callback_url when a public HTTPS receiver can validate and idempotently process status changes.