Skip to main content
Splits text hierarchically using multiple separators (paragraphs → sentences → words). Ideal for structured documents like markdown.

​
Quick Start

from praisonaiagents import Agent

agent = Agent(
    instructions="Answer questions from documentation.",
    knowledge={
        "sources": ["docs/"],
        "chunker": {
            "type": "recursive",
            "chunk_size": 512
        }
    }
)

response = agent.start("How do I configure the settings?")

​
When to Use

Good For

  • Markdown documentation
  • Technical manuals
  • Structured content
  • Code with comments

Consider Alternatives

  • Unstructured prose
  • Stream of consciousness
  • Very short documents
  • Topic-based splitting needed

​
Parameters

ParameterTypeDefaultDescription
chunk_sizeint512Max tokens per chunk
tokenizer_or_token_counterstr"gpt2"Tokenizer for counting

​
Examples

​
Documentation

agent = Agent(
    instructions="Help users find answers in docs.",
    knowledge={
        "sources": ["README.md", "docs/"],
        "chunker": {
            "type": "recursive",
            "chunk_size": 512
        }
    }
)

​
Large Codebase

agent = Agent(
    instructions="Explain code and architecture.",
    knowledge={
        "sources": ["src/"],
        "chunker": {
            "type": "recursive",
            "chunk_size": 1024
        }
    }
)

​
How It Works

The recursive approach tries larger separators first (paragraphs), then falls back to smaller ones (sentences, words) only when needed.

​
Best Practices

  1. Match chunk size to content density - Dense technical docs need smaller chunks
  2. Use with markdown - Recursive chunking respects markdown structure well
  3. Combine with semantic search - The hierarchical splits provide logical boundaries for retrieval