Custom Chunking

Custom chunking allows you to implement your own chunking strategy by creating a class that inherits from ChunkingStrategy. This is useful when you need to split documents based on specific separators, apply custom logic, or handle domain-specific content formats.

1. Create a Python file

   custom_chunking.py

   from typing import List
   import asyncio
   from agno.agent import Agent
   from agno.knowledge.chunking.base import ChunkingStrategy
   from agno.knowledge.content import Document
   from agno.knowledge.knowledge import Knowledge
   from agno.knowledge.reader.pdf_reader import PDFReader
   from agno.vectordb.pgvector import PgVector
   
   class CustomChunking(ChunkingStrategy):
       def __init__(self, separator: str = "---", **kwargs):
           self.separator = separator
   
       def chunk(self, document: Document) -> List[Document]:
           # Split by custom separator
           chunks = document.content.split(self.separator)
   
           result = []
           for i, chunk_content in enumerate(chunks):
               chunk_content = self.clean_text(chunk_content)  # Use inherited method
               if chunk_content:
                   meta_data = document.meta_data.copy()
                   meta_data["chunk"] = i + 1
                   result.append(Document(
                       id=f"{document.id}_{i+1}" if document.id else None,
                       name=document.name,
                       meta_data=meta_data,
                       content=chunk_content
                   ))
           return result
   
   db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
   
   knowledge = Knowledge(
       vector_db=PgVector(table_name="recipes_custom_chunking", db_url=db_url),
   )
   
   asyncio.run(knowledge.ainsert(
       url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf",
       reader=PDFReader(
           name="Custom Chunking Reader",
           chunking_strategy=CustomChunking(separator="---"),
       ),
   ))
   
   agent = Agent(
       knowledge=knowledge,
       search_knowledge=True,
   )
   
   agent.print_response("How to make Thai curry?", markdown=True)
   
2. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   

   uv venv --python 3.12
   .venv\Scripts\activate
   
3. Install dependencies

   uv pip install -U agno sqlalchemy psycopg pgvector
   
4. Run PgVector

   docker run -d \
     -e POSTGRES_DB=ai \
     -e POSTGRES_USER=ai \
     -e POSTGRES_PASSWORD=ai \
     -e PGDATA=/var/lib/postgresql/data/pgdata \
     -v pgvolume:/var/lib/postgresql/data \
     -p 5532:5432 \
     --name pgvector \
     agno/pgvector:16
   
5. Run the script

   python custom_chunking.py
   

Custom Chunking Params

Parameter	Type	Default	Description
separator	str	"---"	The string used to split the document content into chunks.