A portable, production-oriented cache package that makes RAG & LLM applications faster by reusing previously generated answers. Built with a 3-tier cache architecture: exact + semantic Redis, persistent ChromaDB with Cohere reranking, automatic tier promotion, and Celery workers. Install it with: pip install TriCacheLLM-MMA
