千问App迎来更新：上线Wan2.5视频模型

Core Insights - The article highlights the upgrade of Qianwen APP with the integration of the latest Wanxiang model 2.5, enhancing video creation capabilities significantly [1] - Wanxiang 2.5 is one of the few video models in the industry that supports simultaneous audio and video output, showcasing advanced understanding and generation tasks across multiple modalities [1] - The model ranks third globally and first domestically in video generation capabilities according to the authoritative LMArena evaluation [1] Summary by Categories Product Features - Qianwen APP now allows users to create 1080P HD singing and dancing videos with natural body movements and accurate lip-syncing using just a photo and a text input, without the need for templates [1] - The maximum video length supported is 10 seconds, and the app can handle various input types including real human photos, pets, anime characters, cultural relics, and cartoon images [1] Industry Position - Wanxiang 2.5 is recognized as a leading model in the industry, particularly for its audio-video synchronization capabilities, which are rare among mobile AI assistants [1] - The model's performance in generating videos places it in a competitive position within the global market, indicating strong technological advancements [1]