Kling VIDEO 3.0 - Generate Video (AI Model · By )

API Documentation

API Reference

Videos

Generate Video

Supported Model Code

Loading models...

Kling VIDEO 3.0 (kling-v3)

kling-v30 models support this endpoint

kling-v3 is an advanced image-to-video generation model developed by Kuaishou. It offers high-fidelity cinematic output, native 4K-class visual fidelity, and flexible duration from 3 to 15 seconds. It features native audio-visual synchronization, multi-shot storytelling, and strong consistency for subjects across complex movements. It also supports advanced control mechanisms like start-to-end frame guidance, element referencing, and precise camera controls. https://kling.ai/document-api/apiReference/model/imageToVideo

Generate video

https://api.modelstream.ai

POST/v1/video/generations

Authentication

BearerAuth

AuthenticationBearer <token>

All API requests must be authenticated using a Bearer token in the Authorization header. Please ensure your API key is active.Authorization: Bearer sk-xxxxxx

Parameter Location: Header Param

Request Body

application/json

These parameters come from the selected model form_schema. Switching models updates this list and the request example.

Generation Mode

Choose the generation mode. Text to Video: prompt only. Image to Video: upload a first-frame image (optional last frame) as the start/end of the video. Reference to Video: upload reference images to keep subject/character/style consistent; refer to them with <<<image_1>>> in the prompt. Only one mode at a time.

multi_shot?boolean

When on, control each shot separately with multiple storyboard prompts (total up to 15s). When off, use a single prompt.

Example Value: false

negative_prompt?string

Describe content you don't want to see. Max 2500 chars.

mode?string

Generation quality mode. Standard is faster, Pro is higher quality.

Example Value: std

Enum/Options:

Standard Mode (720P): stdProfessional Mode (1080P): pro

duration?number

Video duration in seconds, 3-15. Standard single-shot values are 5 or 10 seconds; multi-shot total must equal the sum of shots, up to 15 seconds.

Example Value: 5Value Range: 3 ≤ value ≤ 15step: 1

cfg_scale?number

Range 0-1, default 0.5. Higher value means stronger relevance to prompt but lower model flexibility.

Example Value: 0.5Value Range: 0 ≤ value ≤ 1step: 0.1

aspect_ratio?string

Output video aspect ratio. For Image-to-Video (first/last frame), the ratio follows the input image and this field is ignored.

Example Value: 16:9

Enum/Options:

16:9 (Landscape): 16:99:16 (Portrait): 9:161:1 (Square): 1:1

watermark_enabled?boolean

When enabled, also produce a watermarked result. Disabled by default.

Example Value: false

Response Parameters

application/json

200apiDocs.responses.successCreateVideoGenTask

task_id?string

Parameter description for Task Id

status?string

Parameter description for Status

400apiDocs.responses.badRequestParams

error?object

Parameter description for Error

message?string

Error Message

type?string

Error Type

param?string

Related Parameters

code?string

Error Code

curl -X POST "https://api.modelstream.ai/v1/video/generations" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kling-v3",
  "prompt": "A cinematic slow-motion shot of a cup of hot coffee with steam swirling gracefully above it, placed on a wooden table next to a rainy window. The raindrops are sliding down the glass, and the warm indoor lighting creates a cozy atmosphere. High quality, photorealistic, 4k.",
  "duration": 5,
  "mode": "std",
  "metadata": {
    "cfg_scale": 0.5,
    "aspect_ratio": "16:9",
    "multi_shot": false,
    "watermark_info": {
      "enabled": false
    }
  }
}'

{
  "task_id": "abcd1234efgh",
  "status": "queued"
}

API Documentation

API Reference

Videos

Generate Video

Supported Model Code

Loading models...

Kling VIDEO 3.0 (kling-v3)

kling-v30 models support this endpoint

Generate video

https://api.modelstream.ai

POST/v1/video/generations

Authentication

BearerAuth

AuthenticationBearer <token>

All API requests must be authenticated using a Bearer token in the Authorization header. Please ensure your API key is active.Authorization: Bearer sk-xxxxxx

Parameter Location: Header Param

Request Body

application/json

These parameters come from the selected model form_schema. Switching models updates this list and the request example.

Generation Mode

multi_shot?boolean

When on, control each shot separately with multiple storyboard prompts (total up to 15s). When off, use a single prompt.

Example Value: false

negative_prompt?string

Describe content you don't want to see. Max 2500 chars.

mode?string

Generation quality mode. Standard is faster, Pro is higher quality.

Example Value: std

Enum/Options:

Standard Mode (720P): stdProfessional Mode (1080P): pro

duration?number

Video duration in seconds, 3-15. Standard single-shot values are 5 or 10 seconds; multi-shot total must equal the sum of shots, up to 15 seconds.

Example Value: 5Value Range: 3 ≤ value ≤ 15step: 1

cfg_scale?number

Range 0-1, default 0.5. Higher value means stronger relevance to prompt but lower model flexibility.

Example Value: 0.5Value Range: 0 ≤ value ≤ 1step: 0.1

aspect_ratio?string

Output video aspect ratio. For Image-to-Video (first/last frame), the ratio follows the input image and this field is ignored.

Example Value: 16:9

Enum/Options:

16:9 (Landscape): 16:99:16 (Portrait): 9:161:1 (Square): 1:1

watermark_enabled?boolean

When enabled, also produce a watermarked result. Disabled by default.

Example Value: false

Response Parameters

application/json

200apiDocs.responses.successCreateVideoGenTask

task_id?string

Parameter description for Task Id

status?string

Parameter description for Status

400apiDocs.responses.badRequestParams

error?object

Parameter description for Error

message?string

Error Message

type?string

Error Type

param?string

Related Parameters

code?string

Error Code

curl -X POST "https://api.modelstream.ai/v1/video/generations" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kling-v3",
  "prompt": "A cinematic slow-motion shot of a cup of hot coffee with steam swirling gracefully above it, placed on a wooden table next to a rainy window. The raindrops are sliding down the glass, and the warm indoor lighting creates a cozy atmosphere. High quality, photorealistic, 4k.",
  "duration": 5,
  "mode": "std",
  "metadata": {
    "cfg_scale": 0.5,
    "aspect_ratio": "16:9",
    "multi_shot": false,
    "watermark_info": {
      "enabled": false
    }
  }
}'

{
  "task_id": "abcd1234efgh",
  "status": "queued"
}