Practical overview
When it fits
Use the repository when you need implementation examples for SQL-backed embeddings and retrieval, or want to compare exact, approximate, hybrid and RAG patterns before adapting one to an application.
Who it is for
- Developers exploring vector search over data held in Azure SQL or SQL Server.
- Teams building retrieval or semantic-search features and comparing SQL-based implementation patterns.
- .NET developers using SQL Client, Entity Framework Core or Dapper, and Python users exploring hybrid search or RAG notebooks.
Documented uses
- Explore common native vector functions in SQL.
- Generate embeddings from OpenAI through Azure SQL and store them in SQL data.
- Retrieve semantically similar text with exact vector similarity search, or use approximate nearest-neighbour search with DiskANN.
- Combine full-text search, BM25 and cosine similarity for hybrid retrieval.
- Study examples of semantic reranking and RAG with Azure SQL and Azure OpenAI.
Requirements & limits
The repository is public and lists an MIT license. The documented samples require an Azure subscription, Azure SQL Database and an Azure OpenAI resource. Cloning locally requires Git LFS; DiskANN specifically requires SQL Server 2025 or Azure SQL.
- This is a code-and-notebook sample repository, not a hosted search or RAG service.
- The examples vary by task: exact vector search, DiskANN approximate search, hybrid search, reranking and RAG are separate sample paths.
- DiskANN has a specific database prerequisite: SQL Server 2025 or Azure SQL.
- The README lists Azure OpenAI as a prerequisite; SQL work can be done with VS Code and the MSSQL extension or with SQL Server Management Studio.
- The repository includes examples across Python, SQL, C#/.NET and Semantic Kernel.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Explore vector functions with the Getting Started notebook
Use the repository’s introductory notebook to become familiar with common vector functions.
Before you start
- An Azure subscription
- An Azure SQL Database
- An Azure OpenAI resource in the subscription
- Visual Studio Code with the MSSQL extension, or SQL Server Management Studio
Steps
Open the repository’s Getting-Started/getting-started.ipynb notebook.
Use the notebook as the introductory sample for common vector functions.
Expected outcome
A guided introduction to common vector functions in Azure SQL and SQL Server.
Guide limitations
- The README identifies this as a simple getting-started sample but does not specify notebook execution instructions.
Explore exact vector search and keyword search
Review the sample showing semantic similarity search over reviews alongside keyword search.
Before you start
- An Azure subscription
- An Azure SQL Database
- An Azure OpenAI resource in the subscription
- Visual Studio Code with the MSSQL extension, or SQL Server Management Studio
Steps
Open the repository’s Vector-Search example.
Review how it uses text vector representations to find reviews contextually similar to a query.
Review how keyword search is integrated to include specified terms in search results.
Expected outcome
An example of combining vector similarity search with keyword search in an SQL database.
Guide limitations
- The README describes the example’s purpose but does not provide detailed setup or execution steps.
Compare options for Vector search
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Azure SQL Vector Search Samples | Explore common native vector functions in SQL. | Cloud, Server | The repository is public and lists an MIT license. The documented samples require an Azure subscription, Azure SQL Database and an Azure OpenAI resource. Cloning locally requires Git LFS; DiskANN specifically requires SQL Server 2025 or Azure SQL. |
| Mettis | Index text documents through an HTTP endpoint. | Linux, Self-hosted, Server |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 14 October 2026
- Original website title
- Azure-Samples/azure-sql-db-vector-search · Official homepage title · retained in its source language