> ## Documentation Index
> Fetch the complete documentation index at: https://docs.jiekou.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Wanx Wan 2.7 Reference-to-Video

Wanx Wan 2.7 reference-to-video model supports multimodal input (text/images/videos). It can use a person or object as the protagonist to generate single-character performance videos or multi-character interaction videos. It supports intelligent storyboarding to generate multi-shot videos. It supports 720P and 1080P resolutions, durations from 2 to 10 seconds, and is billed by the second. The output includes audio by default.

<Tip>
  This is an **asynchronous** API and only returns the asynchronous task's task\_id. You should use this task\_id to request the [Get Task Result API](/en/docs/models/reference-get-async-task-result) to retrieve the generated result.
</Tip>

## Request Headers

<ParamField header="Content-Type" type="string" required={true}>
  Enumerated value: `application/json`
</ParamField>

<ParamField header="Authorization" type="string" required={true}>
  Bearer authentication format: Bearer \{\{API key}}.
</ParamField>

## Request Body

<ParamField body="seed" type="integer">
  Random seed, used to improve the reproducibility of generated results. Value range: \[0, 2147483647].

  Value range: \[0, 2147483647]
</ParamField>

<ParamField body="size" type="string" default="1920*1080">
  Output video resolution (width*height), which affects cost. 720P tier: 1280*720 (16:9), 720*1280 (9:16), 960*960 (1:1), 1088*832 (4:3), 832*1088 (3:4). 1080P tier: 1920*1080 (16:9), 1080*1920 (9:16), 1440*1440 (1:1), 1632*1248 (4:3), 1248\*1632 (3:4).

  Available values: `1280*720`, `720*1280`, `960*960`, `1088*832`, `832*1088`, `1920*1080`, `1080*1920`, `1440*1440`, `1632*1248`, `1248*1632`
</ParamField>

<ParamField body="audio" type="boolean" default={true}>
  Whether to generate a video with sound, which affects cost. Default is true (video with sound).
</ParamField>

<ParamField body="media" type="object[]" required={true}>
  Reference media array, used to extract character appearance, motion, and voice timbre. Corresponds to character1, character2, etc. in the prompt in array order. Number of images: 0–5; number of videos: 0–3; total number does not exceed 5. Image formats: JPEG, JPG, PNG, BMP, WEBP; resolution \[240,8000] pixels; no more than 10 MB. Video formats: MP4, MOV; duration 1–30 seconds; no more than 100 MB. Audio formats: MP3, WAV, FLAC; duration 3–30 seconds.

  Array length: 1 - 5

  <Expandable title="properties" defaultOpen={true}>
    <ParamField body="url" type="string" required={true}>
      Media file URL.
    </ParamField>

    <ParamField body="type" type="string" required={true}>
      Media type. reference\_image: reference image, used to extract character appearance; reference\_video: reference video, used to extract character motion and appearance; first\_frame: first-frame image, controls the starting frame of the video.

      Available values: `reference_image`, `reference_video`, `first_frame`
    </ParamField>

    <ParamField body="reference_voice" type="string">
      Character reference audio URL, used to clone the character's voice timbre and generate a video with sound. Format: MP3, WAV, FLAC; duration 3–30 seconds.
    </ParamField>
  </Expandable>
</ParamField>

<ParamField body="prompt" type="string" required={true}>
  Text prompt, used to describe the elements and visual characteristics expected in the generated video. Use character1, character2, etc. to reference the reference characters. Each reference (video or image) contains only a single character. Chinese and English are supported, up to 1500 characters.

  Length limit: 0 - 1500
</ParamField>

<ParamField body="duration" type="integer" default={5}>
  Duration of the generated video, in seconds, billed by the second. Integer value range: \[2, 10].

  Value range: \[2, 10]
</ParamField>

<ParamField body="shot_type" type="string" default="single">
  Shot type. single indicates a single shot (default), and multi indicates multiple shots. This parameter has higher priority than the prompt.

  Available values: `single`, `multi`
</ParamField>

<ParamField body="watermark" type="boolean" default={false}>
  Whether to add a watermark identifier. The watermark is located in the lower-right corner of the video.
</ParamField>

<ParamField body="negative_prompt" type="string">
  Negative prompt, used to describe content you do not want to appear in the video. Chinese and English are supported, up to 500 characters.

  Length limit: 0 - 500
</ParamField>

## Response

<ResponseField name="task_id" type="string" required={false}>
  Use task\_id to request the [Get Task Result API](/en/docs/models/reference-get-async-task-result) to retrieve the generated output.
</ResponseField>