Chonkie — Fast, Lightweight Text Chunking for RAG

Chonkie is a high-performance Python library for splitting text into chunks optimized for retrieval-augmented generation. It provides 11 specialized chunkers, a composable Pipeline API, and integrations with 30+ embedding providers, vector databases, and LLMs.

When to Use This Skill

Use this skill when users want to:

• Chunk or split text, documents, code, or markdown for RAG
• Build document ingestion pipelines for vector databases
• Choose a chunking strategy (token, sentence, semantic, recursive, code-aware)
• Set up end-to-end workflows: fetch → preprocess → chunk → embed → store
• Use the Chonkie REST API server for chunking-as-a-service
• Process large document batches with async/multiprocessing
• Configure tokenizers, embedding models, or recursive chunking rules

Installation