4.7k stars · MIT · v1.6.8 (2026-06-01) · Track this in Rising ↗
Focused text-chunking library — semantic and recursive splitters that respect message boundaries, so subtopics come out clean.
Radar check
What we found when we opened it. Every bar is one fact, with its date.
Last code pushed 23 Oct 2026, 4 days before this check.
Latest release v1.7.0 on 7 Jul 2026.
Public since March 2025, a little over 1 year.
4,807 stars and 366 forks.
18 commits and 5 new issues in the last 90 days.
49 contributors. 3 pull requests merged in 90 days.
No security policy file says how to report a problem. No published security advisories.
How the bars are scored
- Alive. Days since the last code push: a week or less 100, a month 90, three months 70, six months 50, a year 30, eighteen months 15, older 0.
- Releases. Age of the latest GitHub release, on a similar scale; no bar when the project publishes no releases.
- Maturity. Age of the project: eight years 100, five 90, three 70, two 55, one 40, six months 25, younger 10.
- Adoption. Stars on a log scale: 100,000 stars is 100, 10,000 is 80, 1,000 is 60, 100 is 40.
- Activity. Commits in the last 90 days: 300 or more 100, 100 85, 30 65, 10 45, any 25.
- Community. People who have contributed: 100 or more 100, 30 80, 10 60, 5 40, 2 25, one 10.
- Security. A published security policy 80, none 40; published advisories are listed, not scored.
Checked by Repo Radar on 26 Oct 2026 from the repository on GitHub. Covered in Edition 2.
▶Repo detailsthe review · specs · pros & cons · install
What it does
It splits text into chunks and names eleven ways of doing it, from a plain count of tokens up to strategies that respect sentences, structure, code and tables. It can also refine the result afterwards by merging overlapping pieces or attaching embeddings, which are the numeric representations used for searching by meaning. It exports chunks as JSON or as a dataset, and it can load them straight into a vector database. There is a self-installed REST interface started with chonkie serve, and a Docker route. The README publishes real size figures: a 505 KB package against 1 to 12 MB for the alternatives, and 49 MB installed against 80 to 171 MB, with speed claims of up to 33 times faster on the simplest chunking. It claims 56 languages. What it does not do is state any limitation anywhere in the README, and it does not name a required Python version either.
Why it matters
Who it suits. Anyone building search by meaning over long documents, chat logs or transcripts, where the input has natural boundaries that a blind token count would cut through. It is also the right size for a machine already running other things, which is the point of the published size figures. Skip it if the documents are short enough that one chunk holds each, and skip it if the pipeline already has chunking that works, because this solves one narrow step.
Verdict. Worth two hours, and the honest framing is that it is a useful component rather than a transformation. MIT, read from the file, copyright 2025 Chonkie, which is as clean as licensing gets. Release v1.7.0 is dated 7 July 2026 and code landed on 2 September 2026, so it is current but not busy. Three things to note. The README contradicts itself on how many integrations exist, saying 45 or more in one place and 32 or more in another, which is a small thing that makes the rest of the numbers harder to trust. The strategies that respect meaning need an embedding model, so they are not free in processor terms even though the library itself is tiny. And a company stands behind it with a hosted product, so it is worth watching whether features drift to the paid side. For a chunker that is a dependency of nothing at all, benbrandt/text-splitter is the smaller option.
Summary (Key Points)from the README
Key features
- Many chunkers: Eleven chunkers, from fixed-size token splitting to LLM-driven segmentation, including SentenceChunker, RecursiveChunker, SemanticChunker, and CodeChunker.
- Fast byte-level chunking: FastChunker is SIMD-accelerated and runs at 100+ GB/s, and it is included in the default install.
- Multilingual support: The README claims out-of-the-box support for 56 languages.
- Pipeline from source to store: Chefs and fetchers load and preprocess data, refineries add overlap merging or embeddings, and handshakes ingest chunks into vector databases.
- Broad vector DB coverage: Handshakes exist for ten or more stores, including Chroma, Elasticsearch, MongoDB, pgvector, Pinecone, Qdrant, Weaviate, LanceDB, and Milvus.
- Embedding and LLM choice: Embedding classes cover 16+ providers, and genies cover five LLM providers. OpenAIGenie can also target any OpenAI-compatible API.
- Custom tokenizers: Built-in character, word, and byte tokenizers are available, and you can pass any function that takes a string and returns an integer.
- Self-hosted REST API: The API server exposes chunkers, refineries, and pipelines, with interactive docs at /docs and reusable pipeline configs stored in SQLite.
- Small footprint: The README reports a 505KB wheel and a 49MB installed size.
Ways to install
| Platform or method | What it needs | Effort |
|---|---|---|
| pip | pip | One command for the basic install |
| uv | uv | One command, which the README describes as faster than pip |
| Full install (pip or uv) | pip or uv | One command, installs all optional extras |
| Docker (API server) | Docker | One command to run the REST API |
Important
- Full install: The README says installing everything is not recommended for production; it suggests installing only the chunkers and integrations you need.
- External services: SlumberChunker, TeraflopAIChunker, MistralOCR, embedding providers, and genies call outside APIs, so they depend on those services and their credentials.
- Optional installs: Several embedding providers (Mistral, Together, Nomic, and others) are installed through the `chonkie[catsu]` extra, so check the install name before you assume a separate package.
- Inconsistent counts: The README says "32+ integrations" in the feature list and "45+ integrations" in the integrations section.
- Self-reported benchmarks: Speed and size comparisons (for example, "33x faster than the slowest alternative") come from the project itself, and the README gives no test setup in the summary.
- Semantic chunking cost: The README says semantic chunking is still "10x lighter" than competitors, but the installed size rises when semantic extras are added.
Written by an AI model from the project’s README; not part of the review, and it can be wrong.
- MIT, read from the file, with published size figures rather than adjectives: 505 KB packaged, 49 MB installed.
- Eleven named strategies behind one interface, so the chunking can be changed without rewriting the pipeline.
- A REST interface and a Docker route are included, so it can be a service as well as a library.
- The README contradicts itself on its own integration count, saying both 45 or more and 32 or more.
- It states no limitations at all and names no required Python version.
- The meaning-aware strategies need an embedding model, and a company with a hosted product stands behind the project.
python3 -m venv venv source venv/bin/activate pip install chonkie # Or with the extras actually needed pip install "chonkie[api,semantic,code]" # As a service instead of a library chonkie serve
