⌘K

Wan 3.0 Video

wan30_video

Wan 3.0 Video is an All-in-One video generation model that supports text-to-video, first-frame and first-last-frame image-to-video, and reference-to-video workflows.

API Notes

wan30_video is the standard Wan 3.0 All-in-One model and supports text-to-video, first-frame and first-last-frame image-to-video, and reference-to-video workflows through one model ID

wan30_video can generate videos up to 30 seconds at 30fps; input.prompt and input.media are not validated as required fields by Vtrix, but the upstream model requires at least one of them

input.media reference, file, and link types cannot be combined with first_frame or last_frame in the same request

Authentication

authorization string required

All APIs require authentication via Bearer Token.

Get API Key:

Visit API Key Management Page to get your API Key.

Usage:

Add to request header:

Authorization: Bearer YOUR_API_KEY

Parameters

model string required

Model ID to use for the request

Value: wan30_video

input array required

Vtrix generation request array. Put the model request body in input[0].params

input[0].params object required

Wan 3.0 request body. Put source content under input and generation options under parameters

input object

Source content for the video generation request

input.prompt string

Text prompt describing the expected video. Chinese and English are supported, and excess content is truncated. In reference mode, refer to media in array order with labels such as “image 1”, “video 1”, and “audio 1”

Maximum length: 20000 characters

input.media array

Media inputs for first-frame, first-last-frame, reference, file, or link workflows. Each item contains type and url

  • first_frame: at most 1 image, used strictly as the first video frame
  • last_frame: at most 1 image, used strictly as the last video frame
  • reference_image: at most 10 reference images
  • reference_video: at most 5 reference videos with a combined duration no longer than 15 seconds
  • reference_audio: at most 5 reference audio files with a combined duration no longer than 15 seconds
  • file: at most 1 document and cannot be combined with link
  • link: at most 1 public web page and cannot be combined with file

input.media[].type string

Media item type

Options: first_frame, last_frame, reference_image, reference_video, reference_audio, file, link

input.media[].url string

Publicly accessible media URL or Base64-encoded image data where supported

Image requirements: JPEG, JPG, PNG without transparency, BMP, or WEBP; each side 240 to 8000 pixels; aspect ratio no greater than 8:1; file size no greater than 20MB

Reference video requirements: MP4 or MOV; each video 1 to 15 seconds; each side 240 to 4096 pixels; aspect ratio no greater than 8:1; file size no greater than 100MB

Reference audio requirements: WAV or MP3; each audio file 1 to 15 seconds; file size no greater than 15MB

File requirements: DOCX, DOC, XLSX, XLS, PPTX, PPT, PDF, TXT, KEY, PAGES, NUMBERS, or MD; file size no greater than 100MB; paginated documents support up to 50 pages

parameters object

Video generation options

parameters.resolution string

Output resolution tier. The successful task’s output resolution is used for billing

Options: 1080P, 720P, 480P

Default: 1080P

parameters.ratio string

Output aspect ratio. adaptive selects a ratio based on the input media and request intent

Options: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16

Default: adaptive

parameters.duration integer

Output duration in seconds. Without video input, use an integer from 2 to 30. With video input, the total input video duration plus output duration cannot exceed 30 seconds. Use -1 to let the model select the duration

Default: 5

parameters.audio boolean

Whether the output video contains audio. Enabling or disabling audio does not change the price

Options: true, false

Default: true

parameters.seed integer

Random seed used to improve reproducibility

Range: 0 - 2147483647

parameters.prompt_extend boolean

Whether to use a language model to rewrite and enrich the prompt. This can improve short prompts but increases processing time

Options: true, false

Default: true

parameters.watermark boolean

Whether to add a watermark to the output video

Options: true, false

Default: false

Polling

Since result generation takes time, you need to poll the task status after creating the task.

The initial response only returns information such as the task ID and initial status. The final result must be obtained by polling the task status endpoint using the task ID.

See the examples on the right for polling requests and responses.

Response Format

error object

Error information, only present when status is failed

code string

Error code

message string

Detailed error message

output array

Generation results, only present when status is completed

content array

List of generated resource content

type string

Resource type

Value: image｜video

url string

Processed resource URL

jobId string

Remote task ID

usage object

Usage statistics, only present when status is completed

cost string

Total cost in USD

discount number

Discount amount

metadata object

Metadata information