Build5Nines SharpVector
A lightweight, in-memory vector database library for embedding semantic search into any .NET application, with customizable vector comparison and preprocessing.
Build5Nines.SharpVector is an in-memory vector database library for .NET applications that stores, searches, and manages text data using vector representations, meant to augment a large language model with additional context using the retrieval-augmented generation pattern. It is customizable and extensible, allowing different vector comparison methods, preprocessing steps, and vectorization strategies to be plugged in.
It is aimed at cases where a full database such as Azure Cosmos DB, PostgreSQL with pgvector, Azure AI Search, or Elasticsearch would be more than is needed, offering a lightweight, local, in-memory alternative that can be embedded directly into a .NET application.
It is installed as a NuGet package, written in C#, and released under the MIT license.