跳到主要内容
Vidu检查中

Vidu / Reference to Video Q3 Ad

vidu-reference-to-video-q3-ad

reference to video输出: 视频输入: 文本 + 图像 + audio
正在加载价格…
官方原价 · 未验证

模型能力

Generate advertising video from reference images.

Vidu Q3 Ad · reference video

输入参数

仅列出此模型接受的字段。必填字段必须提供。

字段类型必填默认值可选值与约束
prompt
string必填—

Text prompt describing the desired video content. Supports Chinese and English, max 5000 characters.

≤ 5000 chars

imageUrls
string[]必填—

Reference images (1-7). The model uses subjects in images to generate consistent video. Supports png/jpeg/jpg/webp, min 128x128 pixels, aspect ratio within 4:1, max 50MB each.

≤ 7 items · HTTPS image URL or media:<uuid>

audioUrls
string[]可选—

Optional audio references (1-3). The model uses these audios as reference for subject-consistent video. MP3 only; each audio max 20s, max 50MB per file.

≤ 3 items · HTTPS audio URL or media:<uuid>

duration
integer必填5

Video duration in seconds. Range: 3-16.

3–16

resolution
string必填"720p"

Resolution. Default: 720p. Options: 720p, 1080p.

可选值: "720p", "1080p"

aspectRatio
string可选"16:9"

Aspect ratio. Default: 16:9. Supports arbitrary ratios. auto option auto-detects from input.

可选值: "16:9", "9:16", "4:3", "3:4", "1:1", "auto"

audio
string可选"true"

Enable audio-video sync. True outputs video with dialogue and sound effects, false outputs silent video.

可选值: "true", "false"

限制与行为

  • 仅支持异步:先提交任务,再轮询结果。
  • prompt: ≤ 5000 chars
  • imageUrls: ≤ 7 items
  • audioUrls: ≤ 3 items
  • duration: 3–16
  • 媒体输入接受公开 HTTPS URL,或 POST /api/v1/media 返回的 media:<uuid> 引用。