Includes essential vector operations like: VECTOR_DISTANCE , VECTOR_NORM \u00a0VECTOR_NORMALIZE. Read more here<\/a><\/p>\n With the General Availability of Native vector support, Azure SQL is no longer just a place to store and query relational data. It is a platform for building intelligent, context-aware, and production-ready AI applications. Whether you’re enabling semantic search, powering RAG pipelines, or orchestrating multi-agent workflows, everything you need is now available directly inside your database.<\/p>\n Here are some examples of what you can do with it:<\/p>\n For Example : A user visits your internal help portal and types:\n> \u201cHow do I reset my password?<\/span>\u201d<\/p>\n Instead of relying on exact keyword matches, Azure SQL uses vector similarity to retrieve the most relevant documents, even if they use different phrasing like:<\/p>\n These documents are stored in the SQL Database with precomputed vector embeddings. The user\u2019s query is embedded and compared using VECTOR_DISTANCE<\/a> function, returning the closest matches , all within a single SQL query.<\/p>\n The user sees the most helpful articles ranked by semantic relevance, not just keyword overlap improving discoverability and user satisfaction.<\/p>\n Example: You\u2019re building a chatbot that answers questions about your company\u2019s HR policies. A user asks:> \u201cWhat\u2019s the parental leave policy?\u201d<\/span><\/p>\n Here\u2019s how the RAG pipeline works with Azure SQL:<\/p>\n This approach ensures the answer is accurate, up-to-date, and grounded in your company\u2019s actual policies<\/p>\n Conversational Agent Example: Insurance Scenario\n<\/strong>A user asks: >\u201cI\u2019m meeting with John Doe. What can you tell me about him?\u201d<\/span><\/p>\n The system activates a collaborative agentic workflow, where each agent interacts with specific tables in the database:<\/p>\n Each agent has an expert-level view of its domain and knows how to query its respective tables. The orchestrator coordinates their outputs and composes a unified summary for the user\u2014such as John Doe\u2019s ID, address, claims history, and payment status.<\/p>\n This modular design allows the system to scale and respond intelligently to complex, multi-step queries making it a true agentic AI solution built on top of Azure SQL.<\/p>\n \ud83d\udc49 Want to see this in action? Check out this video<\/a><\/p>\n The power of this capability goes far beyond individual patterns. Let\u2019s look at how it fits into the broader AI ecosystem and how our customers are using it<\/p>\n Native vector support in Azure SQL is deeply integrated with the broader Microsoft AI ecosystem\u2014enabling seamless development across tools like Semantic Kernel<\/a>, LangChain<\/a>, EF Core<\/a> and Logic Apps<\/a><\/p>\n Vectors in Azure SQL are stored in an optimized binary format but are exposed as JSON arrays for developer convenience. With GA, supported<\/strong> drivers can now leverage enhancements to the TDS protocol to transmit vector data more efficiently in binary format.\nThis reduces payload size, eliminates the overhead of JSON parsing, and preserves full floating-point precision\u2014resulting in faster, more accurate performance for high-dimensional vector workloads in AI and ML scenarios . More info will be updated here.<\/a><\/p>\n With GA, we\u2019ve removed several limitations that existed during preview. From improved indexing support (preview) to expanded driver compatibility, tighter integration with Azure SQL\u2019s enterprise-grade security model. The GA release is more robust, scalable, and production-ready.\nYou can find the full list of changes and lifted restrictions in our documentation here.<\/a><\/p>\n Mondra, a sustainability intelligence platform, is using\u00a0Azure SQL\u2019s native vector support<\/strong>\u00a0and\u00a0Azure OpenAI<\/strong>\u00a0to power their AI assistant,\u00a0Sherpa<\/em>. This agentic solution helps food retailers analyze the environmental impact of over 60,000 products and 1 million ingredients\u2014reducing lifecycle assessment time from weeks to just\u00a0four hours<\/strong><\/p>\n \u201cWhen we process new data, we have a massive amount that needs to go out immediately. Azure SQL Database Hyperscale was the only tier that allowed us to do a massive data push without affecting the user experience in our platform.\u201d\n\u2014\u00a0Marco De Sanctis, CTO, Mondra<\/em><\/p>\n Sherpa combines\u00a0Vector search<\/strong>,\u00a0Semantic Kernel<\/strong>, and\u00a0agentic orchestration<\/strong>\u00a0to deliver personalized insights, generate reports, and support scenario analysis\u2014all while maintaining performance, scalability, and data privacy. It\u2019s a real-world example of how native vector support in Azure SQL enables intelligent, production AI applications. Read the full story here<\/a><\/p>\n Azure SQL combines enterprise-grade performance with intelligent query optimization. Vector similarity search is executed alongside traditional SQL queries enabling hybrid workloads without compromising speed or scale.\nAnd we\u2019re not stopping there.<\/p>\n DiskANN-Based Vector Index<\/strong><\/p>\n The next evolution in performance is already underway with\u00a0DiskANN<\/strong>, a high-performance, disk-based approximate nearest neighbor indexing algorithm.\nDiskANN-based vector indexing is currently available in\u00a0private preview for Azure SQL<\/strong>\u00a0and in\u00a0public preview for SQL Server 2025<\/strong>. This unlocks fast vector search at scale, even on massive datasets that exceed memory limits.\nRead more here<\/a><\/p>\n New family of built-in AI functions that bring model-driven intelligence directly into your database\nThese are currently in preview<\/a> in SQL Server 2025 and will be coming soon to Azure SQL:<\/p>\n Too much to digest and just getting started? We\u2019ve got you!\nWhether you’re new to Azure, exploring Azure SQL for the first time, or just beginning your journey with Retrieval-Augmented Generation (RAG), this is your moment.\nNative vector support in Azure SQL is designed to be beginner-friendly, developer-ready, and enterprise-tough. You don\u2019t need to be an AI expert to get started.<\/p>\n We\u2019ve built a one-click accelerator<\/strong> <\/a>that automates the entire Azure SQL + Azure OpenAI setup for a RAG application in under 5 minutes, reducing manual configuration to just click and go.\nA Streamlit web UI<\/a> that showcases Azure SQL\u2019s native vector search on sample data (e.g. semantic product search or resume matching) with no coding required .\nNo complex setup. No external dependencies. Just clone, deploy, and explore.<\/p>\n Explore the docs, try the samples, and start transforming your data into intelligence.<\/p>\n![\"Image](\"https:\/\/devblogs.microsoft.com\/azure-sql\/wp-content\/uploads\/sites\/56\/2024\/11\/createvectortable.png\")
<\/a><\/p>\nBuilt-in Vector Functions<\/h3>\n
![\"Image](\"https:\/\/devblogs.microsoft.com\/azure-sql\/wp-content\/uploads\/sites\/56\/2024\/11\/vectordistance.png\")
<\/a><\/p>\n
\nUse Cases<\/h2>\n
\ud83d\udd0d<\/strong> Semantic Search<\/strong><\/h2>\n
\n
\ud83d\udcda<\/strong> Retrieval-Augmented Generation (RAG)<\/strong><\/h2>\n
\n
\ud83e\udd16Conversational Agents<\/strong><\/h2>\n
\n
![\"Screenshot](\"https:\/\/devblogs.microsoft.com\/azure-sql\/wp-content\/uploads\/sites\/56\/2025\/06\/Screenshot-2025-06-18-225119.png\")
<\/a><\/p>\n\n
\nRich Ecosystem Integrations<\/strong><\/h2>\n
![\"Integrations](\"https:\/\/devblogs.microsoft.com\/azure-sql\/wp-content\/uploads\/sites\/56\/2025\/06\/Picture1.png\")
<\/a><\/p>\nEnhancements to TDS Protocol<\/strong><\/h2>\n
GA = Goodbye Limitations<\/strong><\/h2>\n
\n\ud83d\udd26Customer Spotlight:<\/strong><\/h2>\n
\nWhat\u2019s Next:<\/strong><\/h2>\n
New Set of AI Functions<\/strong><\/h3>\n
\n
\n\ud83d\ude80<\/strong>Zero to RAG in under 5 minutes:<\/h2>\n
\ud83d\udca1Dont move your data to AI. Bring AI to your data<\/h2>\n
\n
\n