참조-투-비디오 작업 생성(주체 호출)

POST

/ent/v2/reference2video

• Official docs: https://platform.vidu.cn/docs/reference-to-video
• Create a reference-to-video task with subject invocation; returns task_id or details
• Control generation with request parameters

Authorizations

bearer

Type

HTTP (bearer)

Request Body

application/json

object

Text prompt
Text description for video generation.
Note: Max 2000 characters

Model name. Allowed: viduq3-mix, viduq3, viduq2-pro, viduq2, viduq1, vidu2.0

• viduq3-mix: Strong visual quality, smart cuts, audio-video sync, best balance
• viduq3: Smart cuts, audio-video sync, strong multi-camera consistency
• viduq2-pro: Reference video, video edit, video replace
• viduq2: Strong motion, rich detail
• viduq1: Clear visuals, smooth transitions, stable camera
• vidu2.0: Fast generation

Duration; default depends on model
viduq3: default 5s, range 3–16
viduq2-pro, viduq2: default 5s, range 1–10
viduq1: default 5s, allowed 5
vidu2.0: default 4s, allowed 4

Random seed
If omitted or 0, random; if set, uses that seed

Motion amplitude, default auto; auto, small, medium, large
Note: Not effective on q2, q3 models

Passthrough parameter
Not processed; data transfer only
Note: Max 1048576 characters

Add watermark

• true: add
• false: no watermark
  Note 1: Fixed AI-generated watermark; off by default
  Note 2: Use watermarked_url from query task API

Watermark position:
1: top-left, 2: top-right, 3: bottom-right, 4: bottom-left
Default: 3

Watermark image URL
Default text if omitted: AI-generated content

Metadata JSON string (passthrough). Example:
{
"Label": "your_label","ContentProducer": "yourcontentproducer","ContentPropagator": "your_content_propagator","ProduceID": "yourproductid", "PropagateID": "your_propagate_id","ReservedCode1": "yourreservedcode1", "ReservedCode2": "your_reserved_code2"
}
If empty, Vidu default metadata is used

Callback URL (POST). Vidu sends status updates; body matches query task API.
Status: processing, success (retry x3), failed (retry x3)
Callback signature authentication applies

Audio-video direct output, default false; true or false

• true: output with audio
• false: without
  Note: viduq3 defaults to true

object[]

Required

Subject info; 1–7 subjects, 1–7 images total

Aspect ratio; q2, q3 support any ratio or auto
Default 16:9; 1:1, 9:16, 16:9, 3:4, 4:3, auto

Smart subject library, default false; true or false

• true: use smart subject library
• false: do not

Resolution by model and duration:
viduq3 (3–16s): default 720p; 540p, 720p, 1080p
viduq2-pro, viduq2 (1–10s): default 720p; 540p, 720p, 1080p
viduq1 (5s): default 1080p; 1080p
vidu2.0 (4s): default 360p; 360p, 720p