# Install kytea with Homebrew, MacPorts, Nix

Toolkit for analyzing text, especially Japanese and Chinese. Version 0.4.7 via Homebrew; verified 2026-08-02. Also installable with nix: nix profile install nixpkgs#kytea.

## Install

```sh
sudo av install brew:kytea
```

Additional install commands:

### macOS

- Homebrew (100%):

```sh
brew install kytea
```

  Evidence: local Homebrew formula metadata

- MacPorts (94%):

```sh
sudo port install kytea
```

  Evidence: MacPorts ports tree: textproc/kytea/Portfile from https://api.github.com/repos/macports/macports-ports/git/trees/master?recursive=1

### Linux

- Nix (92%):

```sh
nix profile install nixpkgs#kytea
```

  Evidence: nixpkgs package indexes: pkgs/by-name/ky/kytea/package.nix from https://api.github.com/repos/NixOS/nixpkgs/git/trees/master?recursive=1

## Package facts

- **Package key:** brew:kytea
- **Package manager:** Homebrew
- **Version:** 0.4.7
- **Source summary:** Toolkit for analyzing text, especially Japanese and Chinese
- **Homepage:** <https://www.phontron.com/kytea/>
- **Repository:** <https://github.com/neubig/kytea>
- **Last updated:** 2026-08-02T01:26:52+09:00
- **Generated:** 2026-08-03T19:37:03+00:00

## Executables

- api-example (alias)
- kytea (alias)
- train-kytea (alias)

## Install behavior

- Bottle: not available

## Freshness

- Page generated: 2026-08-03
- Package-manager version: 0.4.7
## Project history and usage

KyTea is the Kyoto Text Analysis Toolkit, a C++ toolkit by Graham Neubig for word segmentation, pronunciation estimation, and other analysis tasks, especially for Japanese and other languages where word or morpheme boundaries are not explicit in raw text.

### Project history

The KyTea NEWS file records version 0.0.1 on 2009-11-05 as the initial release of KyWs and KyPe. The public GitHub repository and README describe the later project as a general text-analysis toolkit focused on Japanese and languages requiring segmentation.

KyTea's design sits in the statistical NLP toolkit tradition before transformer-based tokenizers became dominant. It gives researchers and language-processing users a trainable analyzer with command-line tools and a library API rather than a closed dictionary-only analyzer.

### Adoption history

KyTea's adoption is mostly academic and language-tooling oriented. It appears in package managers such as Homebrew, MacPorts, and Nix because Japanese and Chinese text-processing workflows often need a reproducible segmenter that can be scripted from Unix pipelines.

### How it is used

The package installs kytea for analysis and train-kytea for model training. Users run it to segment text, estimate pronunciations, prepare corpora, or build custom models for domains where default Japanese tokenizers are not enough.

### Why package nerds care

KyTea is a nice package-manager artifact because it packages research NLP into classic Unix form: a small binary, a training command, headers, and docs. It is especially useful in multilingual text-analysis stacks where tokenization is a prerequisite rather than the final task.

### Timeline

- 2009-11-05: KyTea 0.0.1 was released as the initial release of KyWs and KyPe.
- 2009-11-16: The ChangeLog records version 0.0.2 with Shift-JIS support and an Autotools build-system change.
- 2026-07-01: The GitHub releases page for neubig/kytea listed no packaged GitHub releases, so the historical anchor remains the repository NEWS and source documentation.

### Related projects

- Related tools and concepts include Japanese morphological analyzers, Chinese word segmentation, MeCab-style tokenization workflows, statistical NLP, corpus preparation, and trainable language-specific text-analysis pipelines.

### Sources

- <https://github.com/neubig/kytea>
- <https://github.com/neubig/kytea/blob/master/ChangeLog>
- <https://raw.githubusercontent.com/neubig/kytea/master/NEWS>
- <https://www.phontron.com/kytea>
- <https://www.phontron.com/kytea/api.html>
- <https://www.phontron.com/kytea/train.html>


## Security Notes

narrow executable package without higher-risk signals.

- **Geiger risk:** green / low
- narrow executable package without higher-risk signals

## Other Package-Manager Records

- Nix - kytea: normalized package name match | nixpkgs package indexes: pkgs/by-name/ky/kytea/package.nix from https://api.github.com/repos/NixOS/nixpkgs/git/trees/master?recursive=1
- MacPorts - kytea: normalized package name match | MacPorts ports tree: textproc/kytea/Portfile from https://api.github.com/repos/macports/macports-ports/git/trees/master?recursive=1


## Combined YAML source

View the package source record on GitHub. [combined/kytea.yml](https://github.com/mxcl/pkgdb/blob/main/combined/kytea.yml)


## Sources

- pkg.so package database
- Geiger risk classifier
- curated package history
- pkgdb category and tag curation
- external package-manager database matches
- cross-ecosystem install command graph