An Index in R2R represents a vector index structure that optimizes similarity search operations across chunks. Indices are crucial for efficient retrieval in RAG applications, supporting various similarity measures and index types optimized for different use cases.

Indices in R2R provide:

  • Fast similarity search capabilities
  • Multiple index method options (HNSW, IVF-Flat)
  • Configurable similarity measures
  • Concurrent index building
  • Performance optimization for vector operations

Available Endpoints

MethodEndpointDescription
POST/indicesCreate a new vector index
GET/indicesList available indices with pagination
GET/indices/{id}Get details of a specific index
PUT/indices/{id}Update an existing index’s configuration
DELETE/indices/{id}Delete an existing index
Was this page helpful?
Built with