Qwen3 embedding and reranking models for retrieval are now available in Amazon SageMaker JumpStart

AWS announced the availability of Qwen3-VL-Embedding-2B and Qwen3-Reranker-4B models in Amazon SageMaker JumpStart for information retrieval and cross-modal understanding.

• Qwen3-VL-Embedding-2B accepts text, images, screenshots, and videos to generate semantically rich vectors supporting 30+ languages
• Qwen3-Reranker-4B outputs relevance scores for query-document pairs across 100+ languages with user-defined instructions
• Models work together: embedding performs initial recall while reranker refines results in subsequent stage
• Deploy models with few clicks via SageMaker Studio or Python SDK

These models enable customers to build comprehensive search pipelines with multimodal capabilities on AWS infrastructure.