According to Tongyi Big Model, the QwEN3-VL-embedding and QwEN3-VL-Reranker model series were officially released and open-sourced today. These two models are based on QwEN3-VL and are specially designed for multi-modal information retrieval and cross-modal understanding, providing a unified and efficient solution for understanding and retrieving mixed content such as graphics and videos.

According to Tongyi Big Model, the QwEN3-VL-embedding and QwEN3-VL-Reranker model series were officially released and open-sourced today. These two models are based on QwEN3-VL and are specially designed for multi-modal information retrieval and cross-modal understanding, providing a unified and efficient solution for understanding and retrieving mixed content such as graphics and videos.