core
API
Generated with
core/org.llm4s/org.llm4s.rag/RAGConfig

RAGConfig

org.llm4s.rag.RAGConfig
See theRAGConfig companion object
final case class RAGConfig(embeddingProvider: EmbeddingProvider, embeddingModel: Option[String], embeddingDimensions: Option[Int], chunkingStrategy: Strategy, chunkingConfig: ChunkingConfig, fusionStrategy: FusionStrategy, topK: Int, rerankingStrategy: RerankingStrategy, rerankTopK: Int, vectorStorePath: Option[String], keywordIndexPath: Option[String], pgVectorConnectionString: Option[String], pgVectorUser: Option[String], pgVectorPassword: Option[String], pgVectorTableName: Option[String], llmClient: Option[LLMClient], systemPrompt: Option[String], tracer: Option[Tracing], documentLoaders: Seq[DocumentLoader], loadingConfig: LoadingConfig)

Configuration for RAG pipeline.

Uses immutable copy pattern for fluent configuration. All settings have sensible defaults for quick start.

Attributes

Example 
// Minimal configuration
val config = RAGConfig()
 .withEmbeddings(EmbeddingProvider.OpenAI)
// Full customization with SQLite
val config = RAGConfig()
 .withEmbeddings(EmbeddingProvider.OpenAI, "text-embedding-3-large")
 .withChunking(ChunkerFactory.Strategy.Sentence, 800, 150)
 .withRRF(60)
 .withSQLite("./rag.db")
 .withLLM(llmClient)
// Using PostgreSQL with pgvector
val config = RAGConfig()
 .withEmbeddings(EmbeddingProvider.OpenAI)
 .withPgVector("jdbc:postgresql://localhost:5432/mydb", "user", "pass", "embeddings")
Companion
object
Graph
Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all

Members list

Value members

Concrete methods

def failOnLoadError(fail: Boolean): RAGConfig

Set whether to stop on first loading error.

Set whether to stop on first loading error.

Attributes

def inMemory: RAGConfig

Use in-memory storage (default)

Use in-memory storage (default)

Attributes

def keywordOnly: RAGConfig

Use keyword search only (no vector)

Use keyword search only (no vector)

Attributes

def vectorOnly: RAGConfig

Use vector search only (no keyword)

Use vector search only (no keyword)

Attributes

def withBatchSize(n: Int): RAGConfig

Set batch size for embedding operations.

Set batch size for embedding operations.

Attributes

def withChunking(strategy: Strategy): RAGConfig

Configure chunking strategy

Configure chunking strategy

Attributes

def withChunking(strategy: Strategy, config: ChunkingConfig): RAGConfig

Configure chunking strategy with custom config

Configure chunking strategy with custom config

Attributes

def withChunking(strategy: Strategy, size: Int, overlap: Int): RAGConfig

Configure chunking with size and overlap

Configure chunking with size and overlap

Attributes

def withCohereReranking(model: String): RAGConfig

Use Cohere cross-encoder reranking

Use Cohere cross-encoder reranking

Attributes

def withDocuments(loader: DocumentLoader): RAGConfig

Add a document loader for build-time ingestion.

Add a document loader for build-time ingestion.

Documents will be ingested when RAG.build() is called.

Attributes

def withDocuments(path: String): RAGConfig

Add documents from a directory path.

Add documents from a directory path.

Attributes

def withDocuments(loaders: Seq[DocumentLoader]): RAGConfig

Add multiple document loaders.

Add multiple document loaders.

Attributes

def withEmbeddingDimensions(dims: Int): RAGConfig

Override embedding dimensions (auto-detected by default)

Override embedding dimensions (auto-detected by default)

Attributes

def withEmbeddings(provider: EmbeddingProvider): RAGConfig

Configure embedding provider

Configure embedding provider

Attributes

def withEmbeddings(provider: EmbeddingProvider, model: String): RAGConfig

Configure embedding provider and model

Configure embedding provider and model

Attributes

def withEmbeddings(provider: EmbeddingProvider, model: String, dimensions: Int): RAGConfig

Configure embedding provider, model, and dimensions

Configure embedding provider, model, and dimensions

Attributes

def withFusion(strategy: FusionStrategy): RAGConfig

Configure fusion strategy

Configure fusion strategy

Attributes

def withLLM(client: LLMClient): RAGConfig

Configure LLM client for answer generation

Configure LLM client for answer generation

Attributes

def withLLMReranking: RAGConfig

Use LLM-based reranking

Use LLM-based reranking

Attributes

def withLoadingConfig(config: LoadingConfig): RAGConfig

Configure document loading behavior.

Configure document loading behavior.

Attributes

def withParallelism(n: Int): RAGConfig

Set parallelism for document processing.

Set parallelism for document processing.

Attributes

def withPgVector(): RAGConfig

Use PostgreSQL with pgvector extension for vector storage.

Use PostgreSQL with pgvector extension for vector storage.

Connects to localhost:5432/postgres with user "postgres" by default. Uses "vectors" as the default table name.

Attributes

def withPgVector(tableName: String): RAGConfig

Use PostgreSQL with pgvector extension for vector storage.

Use PostgreSQL with pgvector extension for vector storage.

Connects to localhost:5432/postgres with user "postgres" by default.

Value parameters

tableName

Table name for vectors

Attributes

def withPgVector(connectionString: String, user: String, password: String, tableName: String): RAGConfig

Use PostgreSQL with pgvector extension for vector storage.

Use PostgreSQL with pgvector extension for vector storage.

Value parameters

connectionString

JDBC connection string (e.g., "jdbc:postgresql://host:port/database")

password

Database password

tableName

Table name for vectors

user

Database user

Attributes

def withRRF(k: Int): RAGConfig

Use RRF fusion with custom k parameter

Use RRF fusion with custom k parameter

Attributes

def withRerankTopK(k: Int): RAGConfig

Configure number of candidates to rerank

Configure number of candidates to rerank

Attributes

def withReranking(strategy: RerankingStrategy): RAGConfig

Configure reranking strategy

Configure reranking strategy

Attributes

def withSQLite(dbPath: String): RAGConfig

Use SQLite for persistent storage

Use SQLite for persistent storage

Attributes

def withSystemPrompt(prompt: String): RAGConfig

Configure system prompt for answer generation

Configure system prompt for answer generation

Attributes

def withTopK(k: Int): RAGConfig

Configure number of results to return

Configure number of results to return

Attributes

def withTracing(tracer: Tracing): RAGConfig

Enable tracing for cost tracking and observability

Enable tracing for cost tracking and observability

Attributes

def withWeightedScore(vectorWeight: Double, keywordWeight: Double): RAGConfig

Use weighted score fusion

Use weighted score fusion

Attributes

Inherited methods

def productElementNames: Iterator[String]

Attributes

Inherited from:
Product
def productIterator: Iterator[Any]

Attributes

Inherited from:
Product
Generated with