Elastic (NYSE: ESTC) today announced Elasticsearch Vector Database, a new serverless offering purpose-built for large-scale vector search and AI applications. Elasticsearch is already one of the most widely used platforms for vector workloads worldwide and now developers get an optimized database with expert-tuned defaults to build high-quality vector search applications quickly. Developers bring their documents and queries, and Elastic handles the embeddings, tuning, and infrastructure underneath.

Building a vector-based application today means stitching together multiple parts of the retrieval pipeline: chunking documents, setting up and hosting embedding and reranking models, configuring indexes, storing vectors efficiently, wiring query-time embeddings and rerankers, and retrieving documents behind the matches. Each step adds operational overhead as data volumes grow, with most pure-play vector databases adding unpredictable pricing on top.

Elasticsearch Vector Database handles all of this automatically without the runaway costs:

  • The right defaults, already set: Expert-tuned, production-grade defaults determine how vectors are stored, indexed and merged, with optimized instance types built for vector workloads. Developers don’t need weeks of manual tuning to get fast vector search working. A single field type handles indexing, embeddings, and chunking, so users get semantic search without building an embedding pipeline. Hybrid search is built in, allowing developers to combine full-text and vector retrieval in one query.

  • High-quality relevance, out of the box: Vector and keyword search run across text, image and multi-modal vectors on one index. Developers can use their own models or Jina AI embedding and reranking models on managed GPUs through the Elastic Inference Service, with no embedding pipeline or model servers to operate, to achieve best-in-class relevance.

  • Scale to hundreds of billions of vectors with predictable costs: Elasticsearch Vector Database combines optimized instance types and automatic quantization through Elastic’s Better Binary Quantization, which shrinks vector memory by up to 32x while keeping search fast and recall high. Pricing is based on data and search capacity, with no opaque compute units and no charges for background operations.

Together, these optimizations help developers ship applications faster while keeping costs low.

“Developers building AI applications shouldn’t need to become infrastructure engineers to get vector search working,” said Ajay Nair, general manager, Elasticsearch and Platform, Elastic. “Elasticsearch has powered vector workloads at scale for years. Today’s launch takes what we’ve learned from those deployments and puts it behind an experience optimized for RAG, agents and semantic search, all without the infrastructure overhead or bill surprises that come with most vector solutions.”

Availability

Elasticsearch Vector Database is available now on Elastic Cloud Serverless. Start a free trial here, create a new serverless project, and select the Vector Database use case to reach a running vector query in minutes.

Additional Materials

Blog: Elasticsearch Vector Database: Ship in minutes, scale affordably to hundreds of billions

About Elastic

Elastic (NYSE: ESTC) integrates its deep expertise in search technology with artificial intelligence to help everyone transform all of their data into answers, actions, and outcomes. The Elasticsearch Platform, which is the foundation for its search, observability, and security solutions, is used by thousands of companies, including more than 75% of the Fortune 100. Learn more at elastic.co.

Elastic and associated marks are trademarks or registered trademarks of elasticsearch B.V. and its subsidiaries. All other company and product names may be trademarks of their respective owners.

Media gallery

About The Author
Business Wire