Sign in

v1

latestOpenAPI 3.1.0MIT2026-07-2614271305.9 KB
Video Generation

wan2.7-video-edit API

  • WAN2.7 (wan2.7-video-edit) model supports video editing for instruction-based editing and video transfer tasks
  • Required: upload exactly 1 video (video_urls); optionally upload up to 4 reference images (image_urls)
  • Two usage patterns:
    • Instruction-only editing: provide video_urls + prompt (e.g. style change, replacing clothing)
    • Instruction + reference images: provide video_urls + image_urls + prompt (e.g. region replacement using elements from reference images)
  • Audio handling: controlled via keep_original_sound (true keeps original audio; false lets the model handle audio intelligently)
  • Asynchronous processing mode, use the returned task ID to query status
  • Generated video links are valid for 24 hours, please save them promptly
  • Billing: charged based on "input video duration + output video duration"; only successful generations are billed, failed tasks are free
post/v1/videos/generations

Request body

model'wan2.7-video-edit' required

Model name, must be wan2.7-video-edit

promptstring

Text prompt describing the desired editing effect. Supports Chinese and English; each character/letter counts as 1, with overflow auto-truncated. Maximum length 5000 characters

negative_promptstring

Negative prompt describing what should not appear in the video. Supports both Chinese and English. Maximum length 500 characters; overflow is auto-truncated

video_urlsstring[] required

Source video URL array. Exactly 1 video is required

Video limits:

  • Formats: mp4, mov
  • Duration: 2 ~ 10 seconds
  • Resolution: width and height in [240, 4096] pixels
  • Aspect ratio: 1:8 ~ 8:1
  • File size: up to 100MB
image_urlsstring[]

Reference image URL array, up to 4 images. Used for the "instruction + reference image" pattern, e.g. replacing clothing or props with elements from a reference image

Image limits:

  • Formats: JPEG, JPG, PNG (transparency not supported), BMP, WEBP
  • Resolution: width and height in [240, 8000] pixels
  • Aspect ratio: 1:8 ~ 8:1
  • File size: up to 20MB
keep_original_soundboolean

Audio handling for the edited video. Default: false

  • false: Smart mode. The model decides based on prompt content — if the prompt mentions audio elements, it may regenerate audio; otherwise it may preserve the original audio of the input video
  • true: Force-preserve the original audio of the input video; do not regenerate
quality'720p' | '1080p'

Video quality, defaults to 720p

Options:

  • 720p: Standard definition, standard price, this is the default
  • 1080p: High definition, higher price
aspect_ratio'16:9' | '9:16' | '1:1' | '4:3' | '3:4'

Video aspect ratio (optional)

Behavior:

  • Not provided: video is generated using an aspect ratio close to the input video
  • Provided: video is generated using the specified aspect_ratio

Output resolution per quality tier:

Quality16:99:161:14:33:4
720p1280×720720×1280960×9601104×832832×1104
1080p1920×10801080×19201440×14401648×12481248×1648
durationnumber

Video duration in seconds

Rules:

  • Default 0: keep the input video duration; no truncation (only set this field when you want to truncate the input video)
  • When set explicitly, range is 2 ~ 10: the system truncates the input video starting at 0s up to the specified duration
seedinteger

Random seed, defaults to random

Notes:

  • Range: 1 ~ 2147483647
  • Fixing the seed reduces variation when iterating on prompts and improves reproducibility
prompt_extendboolean

Whether to enable intelligent prompt rewriting. When enabled, a large model will optimize the prompt, which significantly improves results for simple or insufficiently descriptive prompts.

Note: Default is false. Omitting the field or sending false will not trigger rewriting; explicitly send true to enable.

callback_urlstring uri

HTTPS callback URL for task completion

Callback Timing:

  • Triggered when task is completed, failed, or cancelled
  • Sent after billing confirmation

Security Restrictions:

  • Only HTTPS protocol is supported
  • Callbacks to internal IP addresses are prohibited (127.0.0.1, 10.x.x.x, 172.16-31.x.x, 192.168.x.x, etc.)
  • URL length must not exceed 2048 characters

Callback Mechanism:

  • Timeout: 10 seconds
  • Up to 3 retries after failure (retries at 1/2/4 seconds after failure)
  • Callback response format is consistent with the task query API response
  • 2xx status codes are considered successful, other status codes trigger retries

Example request

{
  "model": "wan2.7-video-edit",
  "prompt": "Dress the character in flashy clothes and put on the hat from the reference image",
  "negative_prompt": "Blurry, low quality",
  "video_urls": [
    "https://example.com/source.mp4"
  ],
  "image_urls": [
    "https://example.com/ref.jpg"
  ],
  "keep_original_sound": true,
  "quality": "720p",
  "aspect_ratio": "16:9",
  "seed": 42,
  "callback_url": "https://your-domain.com/webhooks/video-task-completed"
}

Response

Video task created successfully

createdinteger

Task creation timestamp

idstring

Task ID

modelstring

Actual model name used

object'video.generation.task'

Specific task type

progressinteger

Task progress percentage (0-100)

status'pending' | 'processing' | 'completed' | 'failed'

Task status

type'text' | 'image' | 'audio' | 'video'

Task output type

Example response

{
  "created": 1757169743,
  "id": "task-unified-1757169743-7cvnl5zw",
  "model": "wan2.7-video-edit",
  "status": "pending",
  "task_info": {
    "can_cancel": true,
    "estimated_time": 120
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_call",
    "credits_reserved": 5,
    "user_group": "default"
  }
}