Meridian.Knowledge 1.0.0

.NET 10.0 This package targets .NET 10.0. The package is compatible with this framework or higher. 
dotnet add package Meridian.Knowledge --version 1.0.0
                    


                    

                
NuGet\Install-Package Meridian.Knowledge -Version 1.0.0
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package. 
<PackageReference Include="Meridian.Knowledge" Version="1.0.0" />
For projects that support PackageReference, copy this XML node into the project file to reference the package. 
<PackageVersion Include="Meridian.Knowledge" Version="1.0.0" />
                    


                            Directory.Packages.props
                        

                    

                
<PackageReference Include="Meridian.Knowledge" />
                    


                            Project file
For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package. 
paket add Meridian.Knowledge --version 1.0.0
                    


                    

                
#r "nuget: Meridian.Knowledge, 1.0.0"
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package. 
#:package Meridian.Knowledge@1.0.0
#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package. 
#addin nuget:?package=Meridian.Knowledge&version=1.0.0
                    


                            Install as a Cake Addin
                        

                    

                
#tool nuget:?package=Meridian.Knowledge&version=1.0.0
                    


                            Install as a Cake Tool

Meridian.Knowledge

Independent open-source project; not affiliated with or endorsed by Microsoft.

Embeddings, a pluggable vector store, and retrieval-augmented generation for .NET — semantic search, hybrid (vector + keyword) search, and a batteries-included RAG pipeline that turns a query straight into a ready-to-send, cited prompt.

Part of Meridian — an opinionated .NET 10 AI framework built on the Microsoft Agent Framework, reachable from modern .NET apps directly and from COM-based applications (VB6, Visual FoxPro, Delphi, VBA) through a COM Callable Wrapper.

Install

dotnet add package Meridian.Knowledge

Requires an IEmbeddingService (ships with this package) backed by an embedding-capable provider — see Meridian.Models.

Example

using Meridian.Knowledge.Embeddings;
using Meridian.Knowledge.Retrieval;
using Meridian.Knowledge.SemanticStore;

// Ingest
var store = new InMemorySemanticStore();
var vector = await embeddingService.EmbedAsync("The customer prefers email over phone calls.");
await store.UpsertAsync(new SemanticRecord("fact-1", vector, "The customer prefers email over phone calls."));

// Hybrid retrieval: vector similarity fused with keyword overlap
var retriever = new HybridRetriever(embeddingService, store, semanticWeight: 0.6, keywordWeight: 0.4);
IReadOnlyList<RetrievedPassage> hits = await retriever.RetrieveAsync("How does the customer like to be contacted?", topK: 5);

// Batteries-included RAG: retrieval -> numbered context -> ready-to-send prompt
var ragBuilder = new RagContextBuilder(retriever);
RagResult result = await ragBuilder.BuildAsync("How should we contact this customer?");
var response = await chatClient.GetResponseAsync(result.Prompt);

What each piece does

Type Role
IEmbeddingService Turns text into vectors via the configured Meridian.Models provider; batch (EmbedManyAsync) and single (EmbedAsync) overloads
ISemanticStore The pluggable vector-store abstraction — upsert, vector search, delete. InMemorySemanticStore ships as the default; implement the interface to back it with Azure AI Search, Qdrant, Postgres/pgvector, or anything else
IRetriever Retrieves the passages most relevant to a natural-language query. Two implementations: SemanticRetriever (pure vector similarity) and HybridRetriever (vector + keyword, weighted fusion)
RagContextBuilder Retrieves, assembles a numbered context block, and templates a ready-to-send prompt — with its sources and a rough token-count estimate attached to the result

Swap SemanticRetriever for HybridRetriever (or vice versa) without touching any calling code — both satisfy IRetriever.

Full documentation

See the Meridian repository for the complete framework — multi-provider models, agents, long-term memory, image generation, and the COM bridge for VB6/Visual FoxPro/Delphi/VBA.

License

MIT © Angel Hernandez

Product Compatible and additional computed target framework versions.
.NET net10.0 is compatible.  net10.0-android was computed.  net10.0-browser was computed.  net10.0-ios was computed.  net10.0-maccatalyst was computed.  net10.0-macos was computed.  net10.0-tvos was computed.  net10.0-windows was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages (2)

Showing the top 2 NuGet packages that depend on Meridian.Knowledge:

Package Downloads
Meridian.Agents

Agent construction, tools, memory and sequential/parallel workflows, built on the Microsoft Agent Framework (MAF).
Meridian.Hosting

The services.AddMeridian(...) front door: wires Models, Knowledge, Agents and Observability into an IServiceCollection with one opinionated, fluent builder.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
1.0.0 125 10/13/2026