UniRouteUniRoute

wan / 2-7-r2v

Commercial use

Wan 2.7 视频 API 是阿里巴巴通义实验室的 AI 视频套件，涵盖四大生成模式——文生视频、图生视频、参考生视频和视频编辑。它支持全模态输入（文本、图像、视频、音频），并输出 720P–1080P 视频。在 UniRoute 通过一个统一 API 即可调用所有四大模式。

Model Type:
Pricing: 低至 $0.112/秒 · 720p
查看完整价格

wan/2-7-image-to-video USD / 秒

规格 / 通道官方云厂商
1080p$0.24$0.168
720p$0.16$0.112

wan/2-7-r2v USD / 秒

规格 / 通道官方云厂商
1080p$0.24$0.168
720p$0.16$0.112

wan/2-7-text-to-video USD / 秒

规格 / 通道官方云厂商
1080p$0.24$0.168
720p$0.16$0.112

wan/2-7-videoedit USD / 秒

规格 / 通道官方云厂商
1080p$0.24$0.168
720p$0.16$0.112

Input

The positive prompt for the generation 55/5000
The negative prompt for the generation 0/500
reference_image
File 1
reference_image 1

File uploads require configured image storage. A public HTTPS image URL can be used directly.

An array of reference image URLs. At least one of the following must be provided: a reference image or a reference video; the total number of images and videos cannot exceed 5.
reference_video
File 1

File uploads require configured image storage. A public HTTPS image URL can be used directly.

An array of reference video URLs. At least one of the reference images and reference videos must be passed in; the total number of images and videos cannot exceed 5
first_frame

File uploads require configured image storage. A public HTTPS image URL can be used directly.

The URL of the first frame image; a maximum of one image can be passed. The aspect_ratio will be automatically ignored after passing the URL, and an approximate aspect ratio video will be generated based on the aspect ratio of the first frame image.
reference_voice

File uploads require configured image storage. A public HTTPS image URL can be used directly.

Audio URL. Used to specify the timbre of the main character in the reference material.
The resolution of output
Output Video Aspect Ratio. If first_frame is not passed: the video will be generated according to the specified aspect_ratio. First_frame has been passed in: Aspect_ratio is automatically ignored, and an approximate aspect ratio video is generated based on the aspect ratio of the first frame.
Output video duration, in seconds. Values ​​must be integers between 2 and 10, with a default value of 5

Enable intelligent prompt rewriting. Enabling it uses a larger model to expand the input prompt, which works better for short prompts, but increases processing time

Add a watermark. The watermark is located in the bottom right corner of the video, and the text is fixed as "AI generated".

The random seed to use for the generation(>=0,<=2147483647)

A configurable parameter. Defaults to true in the Playground.

Output

output typevideoExample output
Download

README

Complete guide to using wan/2-7-r2v