Changelog

Added support for custom embedding models through OpenAI-compatible endpoints

Introduced new analytics dashboard for monitoring RAG performance metrics

Enhanced vector search performance with optimized indexing strategies

Implemented additional encryption layers for data at rest

Launched support for multiple vector store providers including Pinecone and Weaviate

Optimized chunk size detection for better context retrieval

Resolved concurrent request handling issues in high-load scenarios

Initial release of RAGLoop platform

Core RAG functionality with OpenAI integration

REST API endpoints for document ingestion and querying

Basic analytics and monitoring