RAG Visualizer

Play tool to visualize how RAG works

1. Parse→2. Chunk→3. Index→4. Retrieve

Source

STEP 1

Upload any PDF. Its text will be extracted and fed into the pipeline below.

Drop your PDF here

or click to browse

Select File

A short passage about RAG — perfect for exploring the tool

Local Vector Indexing

STEP 3

Each chunk is converted into a numerical vector (embedding) by an AI model running entirely in your browser. No data leaves your device.

← Generate chunks first using the panel below

Interactive Retrieval

STEP 4

Your question is vectorized and compared to all chunks. The closest matches are returned using cosine similarity.

Visual Splitter & Vector Space

STEP 2

Configure how your text is split into chunks. Tune size, overlap, delimiter, or enable semantic splitting (AI-powered). Hit Apply to see the chunks update in real time.

Pipeline Config

1. Segmentation

Delimiter

Character to strictly split on first.

2. Intelligence

Semantic Merging

Sensitivity80%

3. Constraints

Max Chunk Size200 chars

Overlap20 chars

0Chunks Generated

0Avg. Char Length

Upload a document to start chunking...