Package com.google.genkit.plugins.weaviate

Class WeaviateVectorStore

java.lang.Object

com.google.genkit.plugins.weaviate.WeaviateVectorStore

public class WeaviateVectorStore
extends Object

Weaviate vector store implementation for RAG workflows.

Provides vector similarity search using Weaviate's native vector search capabilities. Supports COSINE, L2_SQUARED, and DOT distance measures.

Example usage:

 // Retrieve documents
 RetrieverResponse response = genkit.retrieve("weaviate/my-collection", Document.fromText("What is AI?"),
 		Map.of("limit", 5));

 // Index documents
 genkit.index("weaviate/my-collection", List.of(Document.fromText("AI is artificial intelligence")));
 

Constructor Summary

Constructors

Constructor	Description
WeaviateVectorStore(io.weaviate.client.WeaviateClient client, WeaviateCollectionConfig config, Embedder embedder)	Creates a new WeaviateVectorStore.

Method Summary

Modifier and Type	Method	Description
Indexer	createIndexer()	Creates an Indexer action for this vector store.
Retriever	createRetriever()	Creates a Retriever action for this vector store.
void	ensureCollectionExists()	Ensures the Weaviate collection (class) exists, creating it if configured to do so.
IndexerResponse	index(ActionContext ctx, IndexerRequest request)	Indexes documents into Weaviate with their embeddings.
RetrieverResponse	retrieve(ActionContext ctx, RetrieverRequest request)	Retrieves documents from Weaviate using vector similarity search.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

WeaviateVectorStore

public WeaviateVectorStore(io.weaviate.client.WeaviateClient client,
 WeaviateCollectionConfig config,
 Embedder embedder)

Creates a new WeaviateVectorStore.

Parameters:

client - the Weaviate client

config - the collection configuration

embedder - the embedder to use

Method Details

ensureCollectionExists

public void ensureCollectionExists()
                            throws GenkitException

Ensures the Weaviate collection (class) exists, creating it if configured to do so.

Throws:

GenkitException - if collection creation fails

createRetriever

public Retriever createRetriever()

Creates a Retriever action for this vector store.

Returns:

the Retriever action

createIndexer

public Indexer createIndexer()

Creates an Indexer action for this vector store.

Returns:

the Indexer action

retrieve

public RetrieverResponse retrieve(ActionContext ctx,
 RetrieverRequest request)
                           throws GenkitException

Retrieves documents from Weaviate using vector similarity search.

Parameters:

ctx - the action context

request - the retriever request

Returns:

the retriever response with matched documents

Throws:

GenkitException - if retrieval fails

index

public IndexerResponse index(ActionContext ctx,
 IndexerRequest request)
                      throws GenkitException

Indexes documents into Weaviate with their embeddings.

Parameters:

ctx - the action context

request - the indexer request

Returns:

the indexer response

Throws:

GenkitException - if indexing fails