On February 18, Hikvision announced that based on the Guanlan big model technology system, Hikvision integrated the large-scale multi-modal model of images and texts with embedded intelligent hardware, and released a series of multi-modal large-model text search and storage products - Wensou NVR and Wensou CVR. According to reports, you can retrieve the target image in seconds by entering a sentence or a word. The search range is wide, supporting high-frequency target searches for security scenes such as people, motor vehicles, and non-motor vehicles, as well as feature searches for specific items and abnormal behaviors.