---
id: fasttext-wheel
version: "0.9.2"
license: MIT
license_treatment: permissive
maintenance: abandoned
---

# fasttext-wheel — fasttext Python bindings

License: permissive · Maintenance: abandoned · Downloads: 2.4M/mo

## What it is and what it does

fasttext-wheel is a Python wrapper around Facebook's fastText C++ library for learning word representations and classifying text. It supports two main workflows: unsupervised training (skipgram and CBOW models for word embeddings) and supervised training for text classification. The package depends on pybind11 for C++ bindings, setuptools for installation, and NumPy for numerical operations.

The library trains models on UTF-8 encoded text files and returns model objects that expose learned word vectors, vocabulary, and labels. It includes methods to save/load models, quantize supervised models for compression, and make predictions on new text. However, the package has been abandoned since September 2020 with no ongoing maintenance, meaning it will not receive bug fixes, security updates, or compatibility patches for newer Python or dependency versions.

Use it for:

- Train word embeddings on a corpus to obtain dense vector representations for downstream NLP tasks.
- Build a text classifier by training a supervised model on labeled text data and evaluating precision/recall on test sets.
- Compress trained classification models using quantization to reduce file size for deployment.
- Retrieve word vectors and vocabulary from a trained model for similarity analysis or feature extraction.
- Predict labels for new text using a pre-trained supervised fastText model.

## Worth the install?

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

fasttext-wheel provides Python bindings for fastText, enabling efficient learning of word representations and text classification through unsupervised and supervised training models.

Yes, but with strong caveats. fasttext-wheel is stable and widely used (top 5000 PyPI packages, 2.3M monthly downloads), with no known vulnerabilities and permissive MIT licensing. However, the package is abandoned—last released in 2020 and its repository archived in 2024. Install only if you need fastText specifically and can accept no future maintenance, security patches, or compatibility fixes. For active projects, consider whether a maintained fork or alternative NLP library better suits your risk tolerance.

## Install

pip install fasttext-wheel
uv add fasttext-wheel
poetry add fasttext-wheel

## Installing fasttext-wheel

Before you install: Medium install friction due to C++11 compilation requirements and pybind11 dependency. The package is abandoned (last release 2020-09-03, repository archived as of 2024-03-22), meaning no maintenance or security updates will be provided going forward.

License in practice: MIT license is permissive, allowing commercial and private use with minimal restrictions—you may use, modify, and distribute the package freely provided you include the license notice.

Quickstart:

pip install fasttext-wheel
import fasttext
model = fasttext.train_unsupervised('data.txt', model='skipgram')
print(model['king'])

Requires a C++11-capable compiler and pybind11; UTF-8 encoded text input is mandatory; Python 2.7 or 3.4+ support only.

Verify before relying:

- Whether wheels for Python 3.7–3.9 and 3.11+ are available or if source compilation is required on those versions.
- Current stability and compatibility with modern NumPy and setuptools versions given the 2020 release date.
- Whether the archived repository will accept security patches or if a maintained fork is recommended for production use.

## Package facts

- License: MIT (permissive)
- Python support: unspecified
- Install friction: medium
- Maintenance: abandoned
- Downloads: 2.4M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

word embeddings python, text classification library, fasttext python bindings, word vectors training, sentence classification, unsupervised word learning, skipgram cbow model, nlp, word-embeddings, text-classification

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