kling-kling-v3-omni-video-generation

Models/kling-v3-omni-video-generation

Model Code

kling/kling-v3-omni-video-generation

Model overview

Video generation

Kling Video 3.0 Omni adds omnidirectional reference support for 3-8 second videos or multiple images to anchor characters and elements. It can match original audio and lip motion, improve video consistency and motion, and supports audio-video synchronization and intelligent shot planning.

Company: Kling AIProvider: Alibaba Cloud

Model capabilities

Input modalityT

Output modalityT

Model experience✓

Function calling-

Structured output-

Web search-

Prefix completion-

Context cache-

Batch inference-

Fine-tuning-

Model pricing

Video generation (720P, no reference video)$0.1343/s

Video generation (1080P, no reference video)$0.1791/s

Video generation (720P, silent, no reference video)$0.08955/s

Video generation (1080P, silent, no reference video)$0.1194/s

Video generation (720P, silent, with reference video)$0.1343/s

Video generation (1080P, silent, with reference video)$0.1791/s

Rate limits and context

Maximum input length-

Context length-

Maximum output length-

RPM300

API reference

EndpointPOST /v1/video/generations

SDKOpenAI-compatible

API code examples

Get API Key

import os
import requests

response = requests.post(
    "https://api.wlrouter.com/v1/video/generations",
    headers={
        "Authorization": f"Bearer {os.environ['WLROUTER_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
  "model": "kling/kling-v3-omni-video-generation",
  "prompt": "A calm sunrise over a modern city skyline.",
  "duration": 4,
  "resolution": "720p",
  "parameters": {
    "generateAudio": False
  }
}
)
response.raise_for_status()
print(response.json())