WeChat has introduced a new AI processing feature in its image-sharing interface, currently undergoing gray-scale testing. Users can access tools for photo enhancement, editing, and information extraction directly from the chat toolbar.
The enhancement category includes background replacement, style conversion, and facial beautification. Editing options cover object removal, image expansion, upscaling, and watermarking. Information extraction supports text recognition, translation, and document generation.
Beyond image tools, WeChat has integrated at least 16 AI features powered by the Xiaowei model. These include AI summaries for articles, automated comments for social media posts, and AI-assisted text generation for status updates.
The Xiaowei assistant also supports visual search, allowing users to identify products, translate foreign text, and summarize images. In chat, users can request tone adjustments, translations, or reply suggestions via voice commands.
These features are currently available to a limited group of users during the testing phase. Tencent is also developing larger versions of its WeLM model, including a 617B parameter version, to further expand these capabilities.
