【#Tech24H】On January 8th, Tongyi officially released and open-sourced the Qwen3-VL-Embedding and Qwen3-VL-Reranker model series. Built upon Qwen3-VL, these models are specifically designed for multimodal information retrieval and cross-modal understanding, providing a unified and efficient solution for the comprehension and retrieval of mixed content like images, text, and videos.