Langchain4j Vector Stores Configuration
Pick and configure LangChain4j EmbeddingStore backends—from in-memory through Pinecone, Qdrant, pgvector, and Milvus—with correct add/search API usage in Java.
Overview
LangChain4j Vector Stores Configuration is an agent skill for the Build phase that documents how to configure and use LangChain4j EmbeddingStore integrations across major vector databases.
Install
npx skills add https://github.com/giuseppe-trisciuoglio/developer-kit --skill langchain4j-vector-stores-configurationWhat is this skill?
- Comparison table across 9 store types: setup, performance, scaling, and feature notes
- EmbeddingStore core methods: add, addAll, ids, and embedded payloads
- Store-specific tradeoffs: Pinecone namespaces, Weaviate hybrid, Qdrant gRPC filtering
- Postgres pgvector and MongoDB paths for teams reusing existing data planes
- Neo4j and Milvus called out for graph-plus-vector and large-scale deployments
- 9 vector stores compared in the reference table
- EmbeddingStore documents add, addAll, and id-associated add variants
Adoption & trust: 1.1k installs on skills.sh; 271 GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You are building Java RAG but do not know which LangChain4j vector store fits your hosting model or how to call add and addAll correctly.
Who is it for?
Solo builders using LangChain4j on Spring or plain Java who need a quick matrix plus interface reference before committing to Pinecone, Qdrant, or pgvector.
Skip if: Python LangChain projects, pure frontend search UX, or teams that only need managed OpenAI vector stores without Java.
When should I use this skill?
Configuring LangChain4j vector stores and EmbeddingStore implementations for agent retrieval or RAG APIs in Java.
What do I get? / Deliverables
You select a store from the comparison matrix and implement EmbeddingStore calls with the right performance and filtering features for your agent API.
- EmbeddingStore configuration aligned to a selected backend
- Ingestion code using add/addAll with ids and metadata embeddeds
Recommended Skills
Journey fit
This skill lands when you are wiring retrieval and memory into a Java agent or API during the build phase. Integrations is the right shelf because it is an API reference for connecting LangChain4j to external vector databases, not frontend or growth work.
How it compares
Use as a LangChain4j-specific vector integration reference—not a generic embeddings tutorial or hosted MCP retrieval server.
Common Questions / FAQ
Who is langchain4j-vector-stores-configuration for?
Java developers and indie agent builders standardizing on LangChain4j who need to compare and configure vector store backends.
When should I use langchain4j-vector-stores-configuration?
During Build while integrating retrieval into your backend or agent-tooling service and choosing between in-memory, SaaS, and self-hosted vector stores.
Is langchain4j-vector-stores-configuration safe to install?
It is documentation-only in the skill body; connecting real stores still needs credentials and network access—review the Security Audits panel on this Prism page before production keys.
SKILL.md
READMESKILL.md - Langchain4j Vector Stores Configuration
# LangChain4j Vector Stores - API References Complete API reference for configuring and using vector stores with LangChain4j. ## Vector Store Comparison | Store | Setup | Performance | Scaling | Features | |------------|-------------|-------------|----------------|---------------------| | In-Memory | Easy | Fast | Single machine | Testing | | Pinecone | SaaS | Fast | Automatic | Namespace, Metadata | | Weaviate | Self-hosted | Medium | Manual | Hybrid search | | Qdrant | Self-hosted | Fast | Manual | Filtering, GRPC | | Chroma | Self-hosted | Medium | Manual | Simple API | | PostgreSQL | Existing DB | Medium | Manual | SQL, pgvector | | MongoDB | SaaS/Self | Medium | Automatic | Document store | | Neo4j | Self-hosted | Medium | Manual | Graph + Vector | | Milvus | Self-hosted | Very Fast | Manual | Large scale | ## EmbeddingStore Interface ### Core Methods ```java public interface EmbeddingStore<Embedded> { // Add single embedding String add(Embedding embedding); String add(String id, Embedding embedding); String add(Embedding embedding, Embedded embedded); // Add multiple embeddings List<String> addAll(List<Embedding> embeddings); List<String> addAll(List<Embedding> embeddings, List<Embedded> embeddeds); List<String> addAll(List<String> ids, List<Embedding> embeddings, List<Embedded> embeddeds); // Search EmbeddingSearchResult<Embedded> search(EmbeddingSearchRequest request); // Remove void remove(String id); void removeAll(Collection<String> ids); void removeAll(Filter filter); void removeAll(); } ``` ## EmbeddingSearchRequest ### Building Search Requests ```java EmbeddingSearchRequest request = EmbeddingSearchRequest.builder() .queryEmbedding(embedding) // Required .maxResults(5) // Default: 3 .minScore(0.7) // Threshold: 0-1 .filter(new IsEqualTo("status", "active")) // Optional .build(); ``` ### EmbeddingSearchResult ```java EmbeddingSearchResult<TextSegment> result = store.search(request); List<EmbeddingMatch<TextSegment>> matches = result.matches(); for( EmbeddingMatch<TextSegment> match :matches){ double score = match.score(); // 0-1 similarity TextSegment segment = match.embedded(); // Retrieved content String id = match.embeddingId(); // Unique ID } ``` ## Vector Store Configurations ### InMemoryEmbeddingStore ```java EmbeddingStore<TextSegment> store = new InMemoryEmbeddingStore<>(); // Merge multiple stores InMemoryEmbeddingStore<TextSegment> merged = InMemoryEmbeddingStore.merge(store1, store2); ``` ### PineconeEmbeddingStore ```java PineconeEmbeddingStore store = PineconeEmbeddingStore.builder() .apiKey(apiKey) // Required .indexName("index-name") // Required .namespace("namespace") // Optional: organize data .environment("gcp-starter") // or "aws-us-east-1" .build(); ``` ### WeaviateEmbeddingStore ```java WeaviateEmbeddingStore store = WeaviateEmbeddingStore.builder() .host("localhost") // Required .port(8080) // Default: 8080 .scheme("http") // "http" or "https" .collectionName("Documents") // Required .apiKey("optional-key") .useGrpc(false) // Use REST or gRPC .build(); ``` ### QdrantEmbeddingStore ```java QdrantEmbeddingStore store = QdrantEmbeddingStore.builder() .host("localhost") // Required .port(6333) // Default: 6333 .collectionName("documents") /