RAG parameter sweep experimentation tool — systematically evaluate vector databases, embedding models, chunking strategies, and retrieval methods using any Vector database. Supports both Voyage AI (hosted) and local sentence-transformers models.

evaluation embedding-models hyperparameter-tuning similarity-score rag vector-search pre-evaluation rag-optimization retrieval-optimization no-mcp zero-llm no-agents relevancy-score chunking-strategies

• Updated Jun 9, 2026
• Python

A comprehensive Pre-RAG prototype dashboard for document parsing, multi-method chunking, vector embedding generation, and hybrid search exploration. Built with React and powered by in-browser ML models.

evaluation eval similarity-score rag pre-evaluation embeddings-model no-mcp zero-llm no-agents relevancy-score

• Updated Jun 15, 2026
• TypeScript