RAG Search Playground

What it's for

• Test console to evaluate natural-language queries against the vector store and calibrate retrieval parameters.
• Hybrid search test bench combining dense embedding similarity and BM25 lexical matching.
• Vector store diagnostic tool to inspect scores, dimensions and returned fragments.

For management of bulk document ingestion without running searches, use RAG Source Manager.

Includes

• Vector query testing
• Similarity threshold tuning
• Top k adjustment
• Hybrid dense sparse toggle
• Vector embedding inspection
• Score decomposition
• Preset queries
• Collection selector

Works with

• Radix UI primitives (the registry variant it ships in)
• React 19, Next.js App Router and Tailwind CSS 4, installed with the shadcn CLI
• Your theme tokens, in light and dark mode