Alibaba has just released its latest models: Qwen3-VL-Embedding and Qwen3-VL-Reranker, the core of which is multi-modal information search.

Alibaba launched the Qwen3-VL-Embedding and Qwen3-VL-Reranker models, which support multi-modal information search, including text search for images and image search for videos. This combination improves search efficiency through the two-stage retrieval process of Embedding and Reranking, and supports more than 30 languages. Embedding-8B performs well on MMEB-V2