Using the In-Memory connector (Preview)

Warning

The Semantic Kernel Vector Store functionality is in preview, and improvements that require breaking changes may still occur in limited circumstances before release.

Overview

The In-Memory Vector Store connector is a Vector Store implementation provided by Semantic Kernel that uses no external database and stores data in memory. This Vector Store is useful for prototyping scenarios or where high-speed in-memory operations are required.

The connector has the following characteristics.

Feature Area	Support
Collection maps to	In-memory dictionary
Supported key property types	Any type that can be compared
Supported data property types	Any type
Supported vector property types	ReadOnlyMemory<float>
Supported index types	Flat
Supported distance functions	CosineSimilarity CosineDistance DotProductSimilarity EuclideanDistance
Supports multiple vectors in a record	Yes
IsFilterable supported?	Yes
IsFullTextSearchable supported?	Yes
StoragePropertyName supported?	No, since storage is in-memory and data reuse is therefore not possible, custom naming is not applicable.

Getting started

Add the Semantic Kernel Core nuget package to your project.

dotnet add package Microsoft.SemanticKernel.Connectors.InMemory --prerelease

You can add the vector store to the dependency injection container available on the KernelBuilder or to the IServiceCollection dependency injection container using extension methods provided by Semantic Kernel.

using Microsoft.SemanticKernel;

// Using Kernel Builder.
var kernelBuilder = Kernel
    .CreateBuilder()
    .AddInMemoryVectorStore();

using Microsoft.SemanticKernel;

// Using IServiceCollection with ASP.NET Core.
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddInMemoryVectorStore();

You can construct an InMemory Vector Store instance directly.

using Microsoft.SemanticKernel.Connectors.InMemory;

var vectorStore = new InMemoryVectorStore();

It is possible to construct a direct reference to a named collection.

using Microsoft.SemanticKernel.Connectors.InMemory;

var collection = new InMemoryVectorStoreRecordCollection<string, Hotel>("skhotels");

Key and Vector property lookup

By default the In-Memory Vector Store connector will read the values of keys and vectors using reflection. The keys and vectors are assumed to be direct properties on the data model.

If a data model is required that has a structure where keys and vectors are not direct properties of the data model, it is possible to supply functions to read the values of these.

When using this, it is also required to supply a VectorStoreRecordDefinition so that information about vector dimension size and distance function can be communicated to the In-Memory vector store.

var collection = new InMemoryVectorStoreRecordCollection<string, MyDataModel>(
    "mydata",
    new()
    {
        VectorStoreRecordDefinition = vectorStoreRecordDefinition,
        KeyResolver = (record) => record.Key,
        VectorResolver = (vectorName, record) => record.Vectors[vectorName]
    });

private class MyDataModel
{
    public string Key { get; set; }
    public Dictionary<string, ReadOnlyMemory<float>> Vectors { get; set; }
}

Overview

The In-Memory Vector Store connector is a Vector Store implementation provided by Semantic Kernel that uses no external database and stores data in memory. This Vector Store is useful for prototyping scenarios or where high-speed in-memory operations are required.

The connector has the following characteristics.

Feature Area	Support
Collection maps to	In-memory dictionary
Supported key property types	Any that is allowed to be a dict key, see the python documentation for details here
Supported data property types	Any type
Supported vector property types	list[float | int] | numpy array
Supported index types	Flat
Supported distance functions	Cosine Similarity Cosine Distance Dot Product Similarity Euclidean Distance Euclidean Squared Distance Manhattan Distance Hamming Distance
Supports multiple vectors in a record	Yes
is_filterable supported?	Yes
is_full_text_searchable supported?	Yes

Getting started

Add the Semantic Kernel package to your project.

pip install semantic-kernel

You can create the store and collections from there, or create the collections directly.

In the snippets below, it is assumed that you have a data model class defined named 'DataModel'.

from semantic_kernel.connectors.memory.in_memory import InMemoryVectorStore

vector_store = InMemoryVectorStore()
vector_collection = vector_store.get_collection("collection_name", DataModel)

It is possible to construct a direct reference to a named collection.

from semantic_kernel.connectors.memory.in_memory import InMemoryCollection

vector_collection = InMemoryCollection("collection_name", DataModel)

Coming soon

More info coming soon.