wan2.6-r2v

Text prompt (required)
Used to describe the elements and visual characteristics expected to be included in the generated video, supporting Chinese and English.

Constraint rules:

• Character calculation: Each Chinese character, letter, and punctuation mark counts as 1 character; excess characters will be automatically truncated;
• Length limit: wan2.6-r2v supports a maximum of 1500 characters;
• Character reference: Reference roles are identified by 「character1, character2」 (each reference video contains only a single role), and the model recognizes roles only through this method.

Example value: character1 is happily watching a movie on the sofa
For prompt skills, refer to: Text-to-Video/Image-to-Video Prompt Guide

Negative prompt (optional)
Used to describe content that is not expected to appear in the video frames, to limit the video frame effects.

Constraint rules:

• Supports Chinese and English, with a maximum length of 500 characters; excess characters will be automatically truncated.

Example value: Low resolution, error, worst quality, low quality, incomplete, extra fingers, poor proportions, etc.

Reference video URL array (required)
⚠️ Important: This parameter directly affects costs; for billing rules, please refer to the “Billing and Rate Limiting” document.

Used to extract character images and timbres (if any) to generate videos that conform to reference characteristics, with specific rules:

• Quantity limit: Maximum 3 videos are supported;
• Role mapping: The 1st URL in the array corresponds to character1, the 2nd to character2, and so on;
• Role constraint: Each reference video contains only one role;
• URL protocol: Supports HTTP/HTTPS; local files need to be uploaded to obtain a temporary URL;

Single video constraints:

• Format: mp4, mov;
• Duration: 2～30 seconds;
• Size: No more than 100MB.

Example value: [“https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/xxx.mp4”]

Video resolution (optional)
⚠️ Important: This parameter directly affects costs; cost = unit price (based on resolution) × duration (seconds); under the same model, 1080P costs more than 720P. Please confirm the model price before calling.

Specify the resolution of the generated video in the format of “width * height” (specific values must be filled in, such as 1280 * 720, 1:1 or 720P cannot be filled in).

• Default value: 1920*1080 (1080P)

Video duration (optional)
⚠️ Important: This parameter directly affects costs; cost = unit price (based on resolution) × duration (seconds).

Duration of the generated video, in seconds.

wan2.6-r2v rules:

• Optional values: 5, 10;
• Default value: 5.

Example value: 5

Shot type (optional)
Specifies the shot composition form of the video, with priority higher than the prompt (e.g., if shot_type=single, a single-shot video will be output even if the prompt requires multiple shots).

Optional values:

• single (default): Output a single-shot video;
• multi: Output a multi-shot video.

Description: Can precisely control the video narrative structure, e.g., single for product display, multi for story shorts.

Example value: single

Watermark switch (optional)
Controls whether to add a watermark to the video; the watermark is fixed at the bottom right corner with the text “AI Generated”.

Optional values:

• false (default): Do not add a watermark;
• true: Add a watermark.

Example value: false

Random number seed (optional)
Used to improve the reproducibility of generation results, value range: [0, 2147483647].

Description:

• Automatically generated by the system if not specified;
• Even with the same seed, results cannot be guaranteed to be completely consistent due to model randomness.

Example value: 12345