Vargha-Kh/Langchain-RAG-DevelopmentKit

Supports pluggable vector stores (Qdrant, Milvus, Chroma, Elasticsearch, Neo4j), embeddings models (OpenAI, Ollama, HuggingFace, Mistral), and advanced retrieval techniques including multi-query expansion, contextual compression, and hypothetical document embedding (HyDE). Implements agent architectures like AgenticRAG and AdaptiveRAG with tool integration (Google Search, Python REPL, retrieval tools) for complex reasoning workflows. Includes MongoDB chat history persistence and Cohere reranking for retrieval optimization, with both CLI and Streamlit UI interfaces.

No commits in the last 6 months.