SqlDatabaseVectorSearch and azure-sql-db-vector-search
These are ecosystem siblings—one provides production-ready application architecture (Blazor Web App with Minimal API for RAG), while the other provides foundational code samples demonstrating the same underlying vector search capabilities in SQL Server/Azure SQL.
About SqlDatabaseVectorSearch
marcominerva/SqlDatabaseVectorSearch
A Blazor Web App and Minimal API for performing RAG (Retrieval Augmented Generation) and vector search using the native VECTOR type in Azure SQL Database and Azure OpenAI.
Leverages Semantic Kernel for embedding generation and chat completion with built-in question reformulation and citation tracking from source chunks. Supports multi-format document ingestion (PDF, DOCX, TXT, MD) with configurable text chunking, vector dimension reduction for embedding models, and response streaming via Minimal API endpoints. Uses EF Core for managing vectors in Azure SQL's native VECTOR type with conversation history and detailed token usage tracking across reformulation, embedding, and completion stages.
About azure-sql-db-vector-search
Azure-Samples/azure-sql-db-vector-search
Samples about using vector in SQL Server and Azure SQL
Demonstrates native vector type support in T-SQL with exact and approximate nearest-neighbor search via DiskANN indexing, plus hybrid search combining vector similarity with full-text ranking. Samples cover end-to-end AI integration including embedding generation from Azure OpenAI, semantic reranking, and retrieval-augmented generation patterns. Includes implementations across multiple data access layers (SqlClient, Entity Framework Core, Dapper, Semantic Kernel SDK).
Scores updated daily from GitHub, PyPI, and npm data. How scores work