Quick Start

import dev.langchain4j.data.document.Document;
import dev.langchain4j.data.document.loader.FileSystemDocumentLoader;
import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.EmbeddingStoreIngestor;
import dev.langchain4j.store.embedding.inmemory.InMemoryEmbeddingStore;
import dev.langchain4j.rag.content.retriever.EmbeddingStoreContentRetriever;
import dev.langchain4j.service.AiServices;

import java.nio.file.Paths;

// 1. Load and ingest document
Document doc = FileSystemDocumentLoader.loadDocument(Paths.get("document.pdf"));
EmbeddingStore<TextSegment> store = new InMemoryEmbeddingStore<>();
EmbeddingStoreIngestor.ingest(doc, store);

// 2. Create RAG-enabled assistant
Assistant assistant = AiServices.builder(Assistant.class)
    .chatModel(chatModel)  // Your ChatModel instance
    .contentRetriever(EmbeddingStoreContentRetriever.from(store))
    .build();

// 3. Query with RAG
String answer = assistant.chat("What is this document about?");

// Load multiple documents from directory
List<Document> docs = FileSystemDocumentLoader.loadDocumentsRecursively(
    Paths.get("documents")
);

// Ingest all at once
EmbeddingStore<TextSegment> store = new InMemoryEmbeddingStore<>();
EmbeddingStoreIngestor.ingest(docs, store);

import dev.langchain4j.data.document.splitter.DocumentSplitters;
import dev.langchain4j.model.embedding.EmbeddingModel;

// Create custom ingestor
EmbeddingStoreIngestor ingestor = EmbeddingStoreIngestor.builder()
    .documentSplitter(DocumentSplitters.recursive(500, 50))  // Custom chunk size
    .embeddingModel(customModel)  // Custom embedding model
    .embeddingStore(store)
    .build();

// Ingest with custom configuration
ingestor.ingest(documents);

import dev.langchain4j.rag.content.retriever.EmbeddingStoreContentRetriever;

// Configure retrieval parameters
ContentRetriever retriever = EmbeddingStoreContentRetriever.builder()
    .embeddingStore(store)
    .maxResults(5)           // Return top 5 results
    .minScore(0.7)           // Minimum similarity score
    .displayName("DocRetriever")
    .build();

Assistant assistant = AiServices.builder(Assistant.class)
    .chatModel(chatModel)
    .contentRetriever(retriever)
    .build();

import java.nio.file.FileSystems;
import java.nio.file.PathMatcher;

// Load only PDF files
PathMatcher pdfMatcher = FileSystems.getDefault()
    .getPathMatcher("glob:**.pdf");

List<Document> pdfs = FileSystemDocumentLoader.loadDocumentsRecursively(
    Paths.get("documents"),
    pdfMatcher
);

import dev.langchain4j.store.embedding.IngestionResult;
import dev.langchain4j.model.output.TokenUsage;

IngestionResult result = EmbeddingStoreIngestor.ingest(document, store);
TokenUsage usage = result.tokenUsage();

System.out.println("Input tokens: " + usage.inputTokenCount());
System.out.println("Total tokens: " + usage.totalTokenCount());