Kling AI Video Generation - Overview

Doc version: v1.0.0 | Last updated: 2026-06-11
This platform fully supports the Kling AI official video generation API. Requests and responses are transparently proxied; parameter semantics are identical to the official API.

Capabilities

Kling AI supports the following task types:

Task Type	API Path	Description
Text to Video	`POST /kling/v1/videos/text2video`	Generate video from text prompts
Image to Video	`POST /kling/v1/videos/image2video`	Generate video driven by images
Omni Video	`POST /kling/v1/videos/omni-video`	Omni multimodal video generation with elements, images, videos
Multi-Image to Video	`POST /kling/v1/videos/multi-image2video`	Generate video from multiple reference images
Motion Control	`POST /kling/v1/videos/motion-control`	Motion template-based video generation
Multi-Elements Editing	`POST /kling/v1/videos/multi-elements`	Replace, remove elements in video
Video Extension	`POST /kling/v1/videos/video-extend`	Extend existing video duration
Lip Sync	`POST /kling/v1/videos/advanced-lip-sync`	Audio-driven lip sync video
Avatar	`POST /kling/v1/videos/avatar/image2video`	Image + audio to digital avatar video
Text to Audio	`POST /kling/v1/audio/text-to-audio`	Generate sound effects from text
Video to Audio	`POST /kling/v1/audio/video-to-audio`	Generate matching audio for video
TTS	`POST /kling/v1/audio/tts`	Text to speech
Custom Voices	`POST /kling/v1/general/custom-voices`	Custom voice management
Image Recognition	`POST /kling/v1/videos/image-recognize`	Image content recognition
Element Management	`POST /kling/v1/general/advanced-custom-elements`	Custom element management
Video Effects	`POST /kling/v1/videos/effects`	Template-based video effects

Video generation is asynchronous: after submission, a task_id is returned immediately. Poll the query endpoint until the task succeeds or fails. On success, the response includes a video download URL.

Standard Workflow

Create Task (POST /kling/v1/videos/{endpoint})
        |  Returns task_id + task_status: "submitted"
        v
Query Status (GET /kling/v1/videos/{endpoint}/{task_id})   <- Poll until succeed or failed
        |  Returns task_result.videos[].url
        v
Download Video

Base URL

Path rule: Append the channel prefix /kling after the platform domain, then the official Kling API path.
For example: official https://api-singapore.klingai.com/v1/videos/text2video -> platform https://platform.dataeyes.ai/kling/v1/videos/text2video.

Authentication

All requests authenticate via the HTTP Authorization header with a Bearer token:

Authorization: Bearer {API_KEY}

{API_KEY} is the API key created in the platform console.

Request Headers

Header	Required	Description
`Authorization`	Yes	`Bearer {API_KEY}`
`Content-Type`	Yes	`application/json`

Models

Model	Description
`kling-v3`	Latest flagship, supports Omni multimodal, multi-shot, sound control
`kling-v2-6`	High quality, supports pro/std modes
`kling-v2-5-turbo`	Fast generation, cost-effective
`kling-v2-1-master`	Fine details, rich quality
`kling-v2-master`	v2 series early version
`kling-v1-6`	v1.6 version
`kling-v1`	v1 base version

Supported models may vary by endpoint. Refer to each endpoint document for details.

Task Status

Status	Meaning
`submitted`	Submitted
`processing`	Processing
`succeed`	Succeeded
`failed`	Failed

Response Format

All endpoints use a unified response wrapper:

{
  "code": 0,
  "message": "string",
  "request_id": "string",
  "data": { ... }
}

Field	Type	Description
`code`	Int	Error code, 0 means success
`message`	String	Error message
`request_id`	String	Request ID for tracking
`data`	Object	Business data

Callback

Task creation endpoints support a callback_url parameter. When set, the system sends POST callbacks on task status changes.

00-Overview

Kling AI Video Generation - Overview#

Capabilities#

Standard Workflow#

Base URL#

Authentication#

Request Headers#

Models#

Task Status#

Response Format#

Callback#

Related Documents#