# pytorch/text

Models, data loaders and abstractions for language processing, powered by PyTorch

Repository: https://github.com/pytorch/text
Canonical: https://ross.abutalabs.com/products/text
Homepage: https://pytorch.org/text
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: nlp, data-loader, deep-learning, pytorch, dataset, models
Archived: true
Last push: 2025-09-10T21:39:02+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 41, release rhythm 8, longevity 100
- inputs: {"age_days": 3552, "days_push": 357, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3554, forks 807 (observed 2026-08-28T04:08:09.654627+00:00)

## What it is
TorchText is a PyTorch companion library providing datasets, data loaders, vocabularies, text transforms, and pre-trained models for natural language processing. Development has stopped, with 0.18 (April 2024) as the final stable release.

## Use cases
- load common NLP datasets in PyTorch
- tokenize and preprocess text for deep learning
- build vocabulary and word vector pipelines
- train text classification models with PyTorch
- use pre-trained NLP models in PyTorch

## When to choose
- you need reproducible NLP data pipelines pinned to torchtext 0.18 or earlier
- you are maintaining legacy PyTorch NLP code that already depends on torchtext

## When to avoid
- starting a new NLP project, since development is stopped
- you need actively maintained tooling or newer Python versions beyond 3.11

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, nlp, data-science
- domain: deep-learning, machine-learning
- platform: python
- tags: torchtext, pytorch, text-processing, datasets, tokenization, vocabularies, deprecated, natural-language-processing

## Member repositories
- pytorch/text (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:09.654627+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:34:18.064038+00:00, confidence not recorded.
  - readme: https://github.com/pytorch/text (fetched 2026-08-28T04:08:09.654627+00:00, sha 08285d783fd0)
  - homepage: https://pytorch.org/text (fetched 2026-08-29T09:28:10.367155+00:00, sha 17da3c2ea6cc)
- Data as of 2026-08-30T08:39:29.467469+00:00.