ToAPIsToAPIs
Kling v2.6

Kling v2.6 動画生成

Generate videos with official Kling v2.6 using explicit reference-image and frame-role semantics

POST
/v1/videos/generations

中国本土のユーザー向け: 中国本土のユーザーは https://toapis.cn を API エンドポイント(Base URL)としてご利用ください。本ドキュメント内の例では https://toapis.com を https://toapis.cn に置き換えてください。

  • Async processing mode, returns task ID for subsequent queries
  • Supports text-to-video, image-to-video, explicit first/last frame control, and audio video
  • Supports standard mode (720P) and professional mode (1080P)
  • Professional mode supports automatic audio generation

Authorization

Authorizationstringヘッダー必須

All API endpoints require Bearer Token authentication.

Authorization: Bearer YOUR_API_KEY

リクエストパラメータ

modelstringボディ必須

Video generation model name, fixed as kling-v2-6.

promptstringボディ必須

Text prompt, maximum 2500 characters.

modestringボディデフォルト: std

Generation mode.

  • std - standard mode (720P, silent video only)
  • pro - professional mode (1080P, supports automatic audio generation)
durationintegerボディデフォルト: 5

Video duration in seconds.

Options: 5 or 10

aspect_ratiostringボディデフォルト: 16:9

Video aspect ratio. Common values: 16:9, 9:16, 1:1

reference_imagesstring[]ボディ

Normal reference images.

  • These images are treated as references only
  • They are not automatically converted into first/last frames
  • Use image_with_roles for explicit frame control
image_with_rolesobject[]ボディ

Explicit image-role array for frame control and mixed inputs.

Show image_with_roles object fields
urlstringボディ必須

Publicly accessible image URL.

rolestringボディ必須

Image role.

Supported values:

  • first_frame
  • last_frame
  • reference
  • reference_image

last_frame is only sent when explicitly declared in image_with_roles. The system no longer infers the last frame from reference_images[1].

negative_promptstringボディ

Negative prompt to exclude unwanted content.

audiobooleanボディデフォルト: false

Whether to automatically generate audio.

Available only in mode: "pro".

watermarkbooleanボディ

Whether to add watermark.

Input Rules

Input shapeBehavior
reference_images onlyNormal references
image_with_roles with only first_frame / last_frameFrame control
Both fields used together, or roles include both frame and reference semanticsMixed mode

Examples

Text-to-Video

{
  "model": "kling-v2-6",
  "prompt": "A golden cat running on a sunlit meadow, slow motion, cinematic quality",
  "mode": "std",
  "duration": 5,
  "aspect_ratio": "16:9"
}

Image Reference

{
  "model": "kling-v2-6",
  "prompt": "Animate the referenced portrait with a subtle smile",
  "reference_images": ["https://example.com/reference.jpg"],
  "mode": "std",
  "duration": 5
}

First and Last Frame Control

{
  "model": "kling-v2-6",
  "prompt": "City timelapse transitioning from day to night",
  "image_with_roles": [
    { "url": "https://example.com/day-city.jpg", "role": "first_frame" },
    { "url": "https://example.com/night-city.jpg", "role": "last_frame" }
  ],
  "mode": "pro",
  "duration": 5
}

Mixed Reference and Frame Input

{
  "model": "kling-v2-6",
  "prompt": "Keep the character identity while moving from one scene to another",
  "reference_images": ["https://example.com/character-reference.jpg"],
  "image_with_roles": [
    { "url": "https://example.com/start-scene.jpg", "role": "first_frame" },
    { "url": "https://example.com/end-scene.jpg", "role": "last_frame" }
  ],
  "mode": "pro",
  "duration": 5
}

Pro Mode + Auto Audio

{
  "model": "kling-v2-6",
  "prompt": "Waves crashing against rocks, seagulls circling in the sky, lighthouse in the distance",
  "mode": "pro",
  "duration": 10,
  "audio": true,
  "aspect_ratio": "16:9"
}

Video generation is an async task. Use the Get Video Task ステータス endpoint to query progress and results.