← All models

Wan2.2-S2V-Detect

Published Publisher Alibabawan2.2-s2v-detect

Model Positioning and Uses

Wan2.2-S2V-Detect is an auxiliary image detection model for the Wanxiang digital human video model wan2.2-s2v. It checks whether a person's portrait meets that model's input specifications before video generation. It does not generate video itself; images that pass detection can be submitted with human voice audio to wan2.2-s2v to generate videos of speaking, singing, or performing. (docs.bailian.console.aliyun.com)

Capabilities and Limitations

  • The input is a person's portrait image; the detection API returns results such as whether the image passed detection and whether a human portrait was detected. (alibabacloud.com)
  • The official API documentation requires images to be provided via publicly accessible HTTP/HTTPS links and supports JPG, JPEG, PNG, BMP, and WebP; both width and height must be 400–7000 pixels. The API detects one image per request, and the documentation applies to the China North 2 (Beijing) region. (alibabacloud.com)
  • The official model information states that function calling, web search, context caching, batch inference, and model tuning are not supported. (help.aliyun.com)

Release Date

The date listed in the official Alibaba Cloud Bailian model availability announcement is 2025-08-25. (docs.bailian.console.aliyun.com)

Pricing

The following prices are listed on official pricing pages, all accessed on 2026-11-18; region and platform conditions are noted separately and should not be treated as the same plan with prices directly converted.

Both sites charge for input images detected in successful requests, including images that do not pass detection; failed requests are not charged. The above prices do not include subsequent wan2.2-s2v video generation costs. (docs.bailian.console.aliyun.com)

Official Sources