# facebookresearch/pytext

A natural language modeling framework based on PyTorch

Repository: https://github.com/facebookresearch/pytext
Canonical: https://ross.abutalabs.com/products/pytext
Homepage: https://pytext.readthedocs.io/en/master/
Language: Python
License: NOASSERTION
License Family: other
Archived: true
Last push: 2022-10-17T19:55:31+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2955, "days_push": 1416, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6292, forks 784 (observed 2026-08-28T04:09:41.423477+00:00)

## What it is
PyText is a deep-learning-based NLP modeling framework built on PyTorch, providing extensible abstractions for text classifiers, sequence taggers, and joint intent-slot models with distributed and mixed-precision training support. It is now deprecated, with users directed to migrate to torchtext.

## Use cases
- train text classification models in pytorch
- build named entity recognition sequence taggers
- train joint intent and slot filling models for dialog
- run distributed nlp model training on multiple gpus
- experiment with nlp model architectures quickly
- semantic parsing for task-oriented dialog

## When to choose
- you are maintaining an existing PyText-based pipeline
- you need the reference implementations of the papers it reproduces

## When to avoid
- starting a new NLP project - use torchtext instead
- you need actively maintained software or long-term support
- you need modern transformer-based LLM workflows

## Facets
- artifact type: framework
- maturity: abandoned
- function: machine-learning, nlp, deep-learning, llm-training
- domain: machine-learning, deep-learning
- platform: python
- tags: pytorch, deprecated, text-classification, sequence-labeling, intent-detection, named-entity-recognition, semantic-parsing, facebook-research, natural-language-processing, linux, gpu

## Member repositories
- facebookresearch/pytext (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:41.423477+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 2026-08-29T17:46:17.448731+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/pytext (fetched 2026-08-28T04:09:41.423477+00:00, sha abbb23dacdb3)
- Data as of 2026-08-30T08:39:29.467469+00:00.