Class EmbedderBase
- Namespace
- LMKit.Abstractions
- Assembly
- LM-Kit.NET.dll
Convenience base class for implementing a custom embedding provider with minimal effort.
public abstract class EmbedderBase : IEmbedder, IQueryEmbedder- Inheritance
-
EmbedderBase
- Implements
- Derived
- Inherited Members
Remarks
Derive from this class and override the single batch method GetEmbeddingsAsync(IEnumerable<string>, CancellationToken) plus the ModelId and EmbeddingSize properties. The single-text overload, the query path (which defaults to the passage path), and the synchronous wrappers are all supplied for you. This is the recommended way to plug a remote service or an in-house endpoint into LM-Kit.
Providers that support asymmetric query embedding should override GetQueryEmbeddingsAsync(string, CancellationToken). Providers with additional capabilities (image embedding, pre-tokenized input) may also implement IImageEmbedder or ITokenizingEmbedder directly.
public sealed class MyHttpEmbedder : EmbedderBase
{
public override string ModelId => "my-service/text-embed-v1";
public override int EmbeddingSize => 1024;
public override async Task<float[][]> GetEmbeddingsAsync(
IEnumerable<string> texts, CancellationToken cancellationToken = default)
{
// POST texts to your endpoint and return one vector per input.
}
}
Properties
- EmbeddingSize
Gets the dimension of the vectors this embedder produces.
- ModelId
Gets a stable identifier for the underlying model or deployment.
Methods
- GetEmbeddings(IEnumerable<string>, CancellationToken)
Synchronously generates embedding vectors for a batch of texts.
- GetEmbeddings(string, CancellationToken)
Synchronously generates the embedding vector for a single text.
- GetEmbeddingsAsync(IEnumerable<string>, CancellationToken)
Asynchronously generates embedding vectors for a batch of texts.
- GetEmbeddingsAsync(string, CancellationToken)
Asynchronously generates the embedding vector for a single text.
- GetQueryEmbeddings(string, CancellationToken)
Synchronously generates the embedding vector for a search query.
- GetQueryEmbeddingsAsync(string, CancellationToken)
Asynchronously generates the embedding vector for a search query, applying any query-specific instruction the model supports.