Today, we are announcing significant changes to Azure AI Search to support customers building production-ready generative AI applications. Azure AI Search significantly increases storage capacity and vector index size at no additional cost, allowing customers to run search augmented generations (RAGs) at any scale without sacrificing cost or performance.
This post explains how customers can:
- Today's changes give you more scalability at lower cost.
- Trust Azure AI Search to handle RAG workloads at scale.
- Apply advanced search strategies to navigate complex data and innovate in ways never before imagined.
Azure AI search
Deliver accurate, hyper-personalized responses with generative AI applications
Announcing low-cost scalability and performance improvements with Azure AI Search
Azure AI Search significantly increases vector and storage capacity, giving customers greater scalability, higher performance, and more data per dollar.
We've increased the amount of capacity and compute available to Azure AI Search Basic and Standard tiers in some regions.
What this means is:
The user is Now you can see up to:
- Vector index size increased by 11x.
- 6x increase in total storage.
- 2x faster indexing and query throughput.
These changes will enable our customers to deliver high-quality experiences for all users and interactions at any scale. Customers can scale their generated AI applications to billions of vector indexes in a single search instance without sacrificing speed and performance.
massive support RAG-based applications with a reliable enterprise search system
More than half of Fortune 500 companies trust Azure AI Search to manage their mission-critical enterprise search and generative AI applications. OpenAI, Otto Group, KPMG, and PETRONAS use Azure AI Search to support search augmented generation (RAG) workloads.
When OpenAI announced “GPT” and the Assistant API powered by RAG at OpenAI DevDay 2023, they needed to ensure their search system could handle unprecedented demand and scale. OpenAI turned to Azure AI Search for its ability to support his large internet-scale RAG workloads.
Azure AI Search now supports RAG functionality for ChatGPT, GPT, and Assistant APIs to provide search capabilities for products such as GPT Store. Every time you want to search for or add a file in one of these products, Azure AI Search is the search system that makes it happen.
As of November 2023, ChatGPT alone 100 million visitors every week,over 2000000 Developers use that API to build. Within two months of announcing custom GPTs, 3 million GPTs were created. These are huge numbers and users are all over the world. It's a truly massive RAG.
Build better applications with the latest advanced search system
Professional services, healthcare, and telecommunications teams realized that using just one search technique, like vector search, just doesn't cut it when it comes to building generative AI applications that behave as designed.
Different retrieval techniques perform better in different use cases. A high-quality search system combines multiple techniques to cover a variety of scenarios that tend to occur in a given application.
With Azure AI Search, applications can immediately apply a variety of approaches, such as hybrid search and semantic reranking, so developers can achieve their goals faster and more effectively.
Telus Health uses advanced RAG to deliver customer care applications
Headquartered in Canada, Telus Health is at the forefront of providing technology-driven solutions and services to employers, individuals, healthcare professionals and insurance companies. The organization has implemented a customer support platform designed to answer users' inquiries about specific health plans and provide guidance for navigating the website. Early implementations relied solely on vector search, which was insufficient to address all requirements within an integrated system. For this reason, Telus Health turned to Azure AI Search, known for its innovative and comprehensive set of search technologies.
Telus Health's Guide team helped design the search strategy and effectively leverage AI search to power the platform. Telus Health expanded its search strategy and implemented hybrid search with semantic reranking to ensure the system efficiently handles both questions about client-specific documents and questions about the use of the organization's website. did. This strategic enhancement, supported by Azure AI Search, significantly improves the accuracy and responsiveness of the platform and validates Telus Health's commitment to providing superior customer support.
NIQ Brandbank enables brands to optimize their online presence with multi-vector search
NIQ Brand Bank supports fast-moving consumer goods (FMCG) brands with solutions that provide rich, relevant content and images on the digital shelf to stay ahead of the competition.
NIQ Brandbank's solution Content Health+ helps brands optimize their online presence with actionable, data-driven guidance and information that shows how their product content compares to competitors in the market. I will make it possible.
With a simple, user-friendly format, this application helps brands improve product placement across retailer search results, increase sales, and increase their online presence.
Content Health+ identifies which product attributes influence organic placement on the digital shelf based on research completed by the NIQ Data Impact team. On the backend, the application uses multi-vector search to search for this finding stored in both images and text, and re-ranks the search to display the most relevant results. This feature provides high-quality recommendations on which content attributes brands should focus on to improve performance and revenue.
For this application to work as designed, Content Health+ was built using hybrid multivector search and semantic ranking. In e-commerce and recommendation applications, combining different search methods brings more ideas and opportunities to life.
“Azure AI Search enables you to use hybrid multivector search to promote the most semantically relevant products to the top using text and image embedding with semantic ranking. From automatic, production-ready data ingestion from Azure data sources to integration with Azure Machine Learning, Azure AI Search is exactly what you need to power your Content Health+ applications.”
—Dr. Gabriel Harris, Principal Data Scientist
Learn more about Azure AI Search here.
These announcements today will make it easier for AI systems to capture information at scale.. Customers can innovate with confidence with Azure AI Search's cutting-edge search technology and enterprise-ready foundation.
For more information about RAG and Azure AI Search, please visit:
