⬡ RAG PIPELINE VISUALIZER
Loading AI models into your browser. Models are cached after first download.
🔢 Embedding model — all-MiniLM-L6-v2 (22 MB)
🔄 Reranker — ms-marco-MiniLM-L-6 (~80 MB)
🤖 LLM — LaMini-Flan-T5-248M (~900 MB)
Initializing…
Node Settings
💬 Chat idle
🤖
Ask anything. Every step of the
RAG pipeline will be visualized live.
🔍 Retrieval & Reranking idle
Retrieved (Top-K)
📚
Retrieved chunks appear here
After Reranking
🔄
Reranked context appears here
🗄️ Vector DB (LanceDB-style) 0 vectors
Embedding…
# Text Preview Vector Preview Source Category Date Added