← All models

Wan2.1-VACE-Plus

Published Publisher Alibabawanx2.1-vace-plus

Model positioning

Wan2.1-VACE-Plus is a unified video creation and editing model in Wanxiang 2.1. It accepts conditional inputs such as text, images, and video, and outputs video. The official announcement lists it as a video generation model. (help.aliyun.com)

Capabilities and uses

  • Supports local video editing, video repainting, frame expansion, video duration extension, and image-reference generation.
  • Can be used to modify or replace local elements in videos, expand frame boundaries, and generate videos with reference to the content and dynamic features of images or videos. (help.aliyun.com)

Limitations

  • The 2.1 API uses asynchronous calls and requires the editing function to be specified through function. Input material requirements vary by function; for example, multi-image reference uses a maximum of 3 reference images.
  • The final output of the video extension function has a total duration of 5 seconds; it does not add 5 seconds to the original video. Consult the API documentation for format, size, and duration limits for the selected function.
  • Official model information states that function calling, structured output, web search, context caching, batch inference, and fine-tuning are not supported. (help.aliyun.com)

Release date

May 14, 2025, as stated in the official model release and update announcement. (help.aliyun.com)

Official pricing

The following are pay-as-you-go video generation prices for 720P output, all retrieved on November 18, 2026. Input is not charged; output is charged by the second of video. Failed requests are not charged. Quotes for different sites and regions are not for the same package and cannot be directly converted.

Other official sources