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

AWS has made Qwen3-VL-Embedding-2B and Qwen3-Reranker-4B available in Amazon SageMaker JumpStart. The embedding model handles multimodal inputs (text, images, screenshots, videos) across 30+ languages, generating semantic vectors for tasks like image-text retrieval and visual question answering. The reranker model takes query-document pairs and outputs relevance scores for text, code, and multilingual content across 100+ languages. Both models are designed to work together in a two-stage retrieval pipeline and can be deployed via SageMaker Studio or the SageMaker Python SDK.

Source: https://aws.amazon.com/about-aws/whats-new/2026/07/qwen3-search-retrieval-on-sagemaker-jumpstart. 8 Sync News only summarizes and links out; content copyright belongs to the authors and original sources.