# RUCAIBox/TextBox

TextBox 2.0 is a text generation library with pre-trained language models

Repository: https://github.com/RUCAIBox/TextBox
Canonical: https://ross.abutalabs.com/products/textbox
Homepage: https://github.com/RUCAIBox/TextBox
Language: Python
License: MIT
License Family: permissive
Topics: text-generation, natural-language-processing, deep-learning, pretrained-models, python, pytorch, seq2seq, natural-language-generation
Last push: 2023-07-27T14:39:30+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2124, "days_push": 1133, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1097, forks 115 (observed 2026-08-28T04:03:34.678124+00:00)

## What it is
TextBox 2.0 is a Python/PyTorch library providing a unified pipeline for applying pre-trained language models to text generation tasks. It bundles 13 generation tasks, 83 datasets, and 47 pre-trained models with training strategies like distributed data parallel.

## Use cases
- benchmark pretrained language models on text generation tasks
- fine-tune seq2seq models for translation or summarization
- run story generation or style transfer experiments
- reproduce published text generation results
- train models with distributed data parallel on multiple GPUs
- evaluate generation quality with standard metrics

## When to choose
- you need a standardized framework to compare many PLMs across generation tasks
- you want ready access to 83 benchmark datasets and 47 models
- you are doing NLP research on text generation and need reproducible pipelines

## When to avoid
- you need production text generation serving rather than research experimentation
- you want the latest actively developed tooling - the last release was mid-2023
- you need a lightweight solution - installation pulls a modified transformers and extra dependencies

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, nlp, llm-training, benchmarking
- domain: deep-learning, large-language-models
- platform: python
- tags: text-generation, pytorch, pretrained-language-models, seq2seq, natural-language-generation, research-framework, natural-language-processing, linux, gpu

## Member repositories
- RUCAIBox/TextBox (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:34.678124+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 2026-08-30T06:46:22.862847+00:00, confidence not recorded.
  - readme: https://github.com/RUCAIBox/TextBox (fetched 2026-08-28T04:03:34.678124+00:00, sha 68f9164ead9a)
  - homepage: https://github.com/RUCAIBox/TextBox (fetched 2026-08-29T12:49:51.829962+00:00, sha faebb024fb62)
- Data as of 2026-08-30T08:39:29.467469+00:00.