---
id: chonkie
version: "1.7.0"
license: MIT License Copyright (c) 2025 Chonkie Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in… (full text in the JSON record)
license_treatment: permissive
maintenance: active
---

# chonkie — 🦛 CHONK your texts with Chonkie ✨ - The no-nonsense chunking library

License: permissive · Maintenance: active · Downloads: 1.3M/mo

## What it is and what it does

Chonkie is a text chunking library designed to prepare documents for retrieval-augmented generation (RAG) systems. It provides multiple chunking strategies—including recursive, semantic, token-based, code-aware, and neural approaches—each suited to different content types and use cases. The library integrates with tokenizers, embedding models, vector databases, and LLMs, allowing you to build end-to-end pipelines that fetch, chunk, refine, embed, and load data into your RAG infrastructure.

The package emphasizes minimal dependencies by default: the base install includes only what's needed for common chunking tasks, with optional extras for specialized features like semantic chunking, code analysis, or API server deployment. It supports both synchronous and asynchronous workflows, can run as a self-hosted REST API, and handles text preprocessing through pluggable "Chef" components for markdown, tables, and OCR.

Use it for:

- Split long documents into token-bounded chunks for LLM context windows before embedding and retrieval
- Chunk code repositories by syntactic structure to preserve function and class boundaries for code search
- Build a RAG pipeline that chains recursive chunking, semantic refinement, and embedding in a single workflow
- Run Chonkie as a microservice API to chunk documents from multiple applications without duplicating logic
- Process markdown or CSV files into structured chunks with context overlap for better retrieval quality

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Chonkie splits text into semantically meaningful chunks for RAG pipelines, offering multiple chunking strategies (recursive, semantic, token-based, code-aware) plus refinement and embedding integration.

Yes. Chonkie is actively maintained, has no known vulnerabilities, uses a permissive MIT license, and offers low install friction with a focused set of dependencies. It solves a genuine problem in RAG pipelines—text chunking—with multiple strategies and integrations. The modular design lets you install only what you need, and the API server option adds deployment flexibility. Suitable for production RAG systems.

## Install

pip install chonkie
uv add chonkie
poetry add chonkie

## Installing chonkie

Before you install: Low friction install with a pure-Python wheel and six runtime dependencies. Active maintenance with recent commits and 4673 repository stars. Supports Python 3.10 through 3.13.

License in practice: MIT License permits unrestricted use, modification, and redistribution in commercial and private projects with minimal obligations—only attribution and license inclusion required.

Quickstart:

pip install chonkie

from chonkie import RecursiveChunker

chunker = RecursiveChunker()
chunks = chunker("Your text here")
for chunk in chunks:
    print(chunk.text)

Requires Python 3.10 or later.

Verify before relying:

- Whether all 45+ integrations mentioned in the description are available in the base install or require optional extras
- Performance characteristics of different chunkers (e.g., the claimed '100+ GB/s' for FastChunker)
- Multilingual support coverage across the stated 56 languages

## Package facts

- License: MIT License Copyright (c) 2025 Chonkie Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 1.3M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

text chunking for RAG, semantic document splitting, token-based text segmentation, chunking library, retrieval augmented generation preprocessing, code chunking, document preparation for embeddings, rag-pipeline, document-chunking, nlp

[View on SkillFed](https://skillfed.io/packages/chonkie) · [View on PyPI](https://pypi.org/project/chonkie/)