New users get a 7-day free trial!

Elasticsearch is the most widely deployed, open source vector database.

With Elasticsearch you won’t have any gaps because it was built to be a complete vector database solution. Build the future of tech and GenAI with:

Hybrid search that combines the precision of keyword search and the context understanding of semantic search with the flexibility of vector search - so you get the best relevance on any data.
Advanced search relevance techniques to return the right results every time: filters, embedding models, inference models, reranking models and more.
Inference APIs that execute native inference with popular LLMs or built-in models for vector embeddings with no external machine learning infrastructure needed.
MCP (Model Context Protocol) that enables you to chat with all of your data in Elasticsearch so you get instant, accurate answers with no querying required

Whether you’re building a conversational search app or experimenting with agentic AI, RAG, or quantization, you don’t need separate tools to build each aspect of your operation.

Take advantage of Elastic Cloud Serverless – the fastest way to start and scale security, observability, and search solutions without managing infrastructure. Built on the industry-first Search AI Lake architecture — which relies on Azure Blob Storage — it combines vast storage, compute, low-latency querying, and advanced AI capabilities to deliver uncompromising speed and scale. Users can choose from Elastic Cloud Hosted and Elastic Cloud Serverless during deployment. Try the new Serverless calculator for price estimates.

Learn more about vector, hybrid, and semantic search at Search Labs and why Elastic won the 2024 US Microsoft Partner of the Year Award.

Other Elastic Offerings
Elastic Observability
Elastic Security

Getting Started
Elastic on Microsoft Azure gives you the ability to deploy Elasticsearch directly within the Azure portal. Elasticsearch resources can be managed in the Azure portal just like any other native Azure service like VMs. NOTE: You need to be configured as a Subscription Owner or Contributor in order to deploy Elasticsearch. When you purchase and manage your Elastic subscription directly from the Azure Marketplace, you can frictionlessly provision your deployment using configuration defaults or using custom configurations—your choice. Deploy in any of the Elastic supported Azure regions. Easily ship and ingest data to Elastic leveraging integrations with popular Azure and third-party services. Analyze and visualize it all with powerful Kibana dashboards.

Billing
Your Azure Marketplace purchase of Elastic will be included in your monthly consolidated billing statement. Your investment in Elastic draws against your committed spend with Microsoft. Elastic Cloud billing is based on your resource consumption metered across three separate dimensions: Deployment capacity, Data Transfer, and Storage. You can monitor and analyze your Elastic spend from the billing and usage page within the Elastic Cloud Console. For more information, visit these links:

Resource-based Pricing
Billing Dimensions

Support

Visit elastic.co/support to find out more about Elastic support. If you are a customer, please go to the Elastic Support Hub to raise a case.

Azure Machine Learning extension

作者 Microsoft

Azure Machine Learning extension allows you to run machine learning workloads on your cluster from on-premises, and across clouds, with seamless Azure Machine Learning experience.