Example 4: A Basic RAG

This workflow includes:

• Retrieve steps: a → b → c → d
• Then, d and e are merged into f
• f and e generate g

🟢 Execution:

1. The first layer runs: a→b → c → d
2. Once both d and e are ready, the edge runs to update f
3. Generate g

🟡 Demo:

1. Create a File block → upload your file → Create a load edge.

2. Run the edge to load the file.

3. You can see the output here:

• Click * to create root node.
• Click + to generate the Embedding View.
• Click Loop Mood Button Icon enables concurrent processing of each element inside a structured block.

4. Create Convert to Text edge.

5. Generate the text.

6. Create a Chunking Edge.

• Auto
  • Easiest and fastest option.
  • Automatically splits content by line breaks.
  • No additional configuration needed.
  • Best for pre-formatted content like lists or short paragraph.

• By length
  • Splits content strictly by character count without using any semantic rules.
  • Configurable:
    • Chunk Size: Max characters per chunk.
    • Overlap: Repeated characters between chunks.
    • Handle Half Word: Prevents breaking words in half .(set to True if using English or similar languages)
  • Ideal when uniform chunk length is needed.

• By character
  • Uses custom user-defined delimiters to split content.
  • Configurable:
    • Delimiters: Choose one or more. (e.g., comma, period, space)
  • Suitable for data with predictable formatting, like CSV or semicolon-separated values.

7. Run the Chunking Edge and create Retrieve Edge.

8. Create a Text Block to describe your question and connect to the Retrieve edge.

9. Modify your Retrieve Edge:

• Query: The text or structured query from user or workflow
• Data source with Vector indexing: The indexed structured content (chunks with embedding vectors)
• Settings:
  • Result Number: Number of top results to return (default: 5)
  • Threshold: Minimum similarity score (0–1). Default is 0.7. Results below this value will be filtered out.

10. Run the Retrieve Edge.

11. Generate the Answer.