Kling AI/kling-v3-omni-video
Kling Video 3.0 Omni is Kuaishou's unified multimodal video model for text-to-video, image-to-video, reference-based generation, video editing, native audio, and multi-shot storytelling up to 15 seconds.
Output: $0.126 / use or 7 uses / $1
Input
prompt * string
Optional text prompt for generation or editing. The official limit is 3072 characters and Kling recommends no more than 2500. Multi-shot prompts can describe 1 to 6 shots, with each shot lasting at least 1 second and all shot durations totaling duration.
first_frame image
Optional first-frame image. Accepts JPG, JPEG, or PNG via URL or base64, up to 50 MB, with both dimensions at least 300 px and an aspect ratio from 1:2.5 to 2.5:1.
last_frame image
Optional last-frame image. Requires first_frame; last-frame-only generation is not supported. Accepts JPG, JPEG, or PNG via URL or base64, up to 50 MB, with both dimensions at least 300 px and an aspect ratio from 1:2.5 to 2.5:1.
refer_image list
Optional reference images for scenes, styles, or visual consistency. Each image must be JPG, JPEG, or PNG via URL or base64, no larger than 50 MB, at least 300 px on both sides, and within a 1:2.5 to 2.5:1 aspect ratio. Official combined image and element limits depend on the other supplied references.
Click to upload images
Support drag and drop, select multiple images
feature_video video
Optional feature-reference video supplied by URL. Accepts MP4 or MOV from 3 to 15.5 seconds, 24 to 60 fps, up to 200 MB, with each side from 700 to 4553 px, total pixels no more than 8294400, and an aspect ratio from 0.4 to 2. Only one reference video is supported; audio must be off when this field is used.
audio enum
Audio behavior: native generates audio matching the visuals, original retains the reference video's original sound, and off produces no audio. Feature-reference video requires off; base-video editing does not support native.
resolution enum
Generated video resolution.
aspect_ratio enum
Generated frame aspect ratio. Required when neither a first frame nor a reference video supplies framing context.
duration int
Generated video duration in seconds, from 3 through 15.
Reset
Output
{
  "task_id": "j2n94b12anrmw0cwdf1bpjde6r",
  "user_id": 1,
  "version": "208a2d1f9892cadc10983fdb7eff9864a8a833126826f0e173cabeeec6af5ec4",
  "error": null,
  "total_time": 549.56,
  "predict_time": 549.55,
  "logs": null,
  "output": [
    "https://vmodel.ai/data/dev/model/kwaivgi/kling-v3-omni-video/tmp8xz0xtiw.mp4"
  ],
  "status": "succeeded",
  "create_at": 1771334410,
  "completed_at": 1771334960,
  "input": {
    "prompt": "A lavish period drama movie trailer set in 18th century Versailles. Opening: a sweeping crane shot over manicured gardens with fountains at golden hour. Cut to a grand ballroom filled with dancers in elaborate silk gowns and powdered wigs, candlelight from enormous chandeliers. A mysterious woman in a crimson dress locks eyes with a nobleman across the room. Quick cuts: secret letters being passed, a duel at dawn in the mist, tears on a beautiful face. Final shot: the woman stands alone on a balcony overlooking Paris at sunset. Elegant gold text fades in: \"KLING V3 OMNI VIDEO\". Sweeping orchestral strings, harpsichord notes, dramatic crescendo.",
    "duration": 15,
    "aspect_ratio": "16:9",
    "multi_shot": true,
    "audio": "native",
    "resolution": "1080p"
  }
}
Generated in: 549.55 seconds
Download
Examples
Pricing
Model pricing for kwaivgi/kling-v3-omni-video. Looking for volume pricing? Get in touch.
When
⚙ model variant is no_video_no_native_audio_720p
$0.126000
per generation at no_video_no_native_audio_720p
or 7 generations for $1
When
⚙ model variant is no_video_no_native_audio_1080p
$0.168000
per generation at no_video_no_native_audio_1080p
or 5 generations for $1
When
⚙ model variant is no_video_no_native_audio_4k
$0.630000
per use
or 1 uses for $1
When
⚙ model variant is no_video_native_audio_720p
$0.168000
per generation at no_video_native_audio_720p
or 5 generations for $1
When
⚙ model variant is no_video_native_audio_1080p
$0.210000
per generation at no_video_native_audio_1080p
or 4 generations for $1
When
⚙ model variant is no_video_native_audio_4k
$0.630000
per use
or 1 uses for $1
When
⚙ model variant is video_input_no_native_audio_720p
$0.189000
per generation at video_input_no_native_audio_720p
or 5 generations for $1
When
⚙ model variant is video_input_no_native_audio_1080p
$0.252000
per generation at video_input_no_native_audio_1080p
or 3 generations for $1
When
⚙ model variant is video_input_no_native_audio_4k
$0.630000
per use
or 1 uses for $1
Readme

Loading...