# google-research/pegasus

Repository: https://github.com/google-research/pegasus
Canonical: https://ross.abutalabs.com/products/pegasus
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2023-07-20T17:45:22+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2352, "days_push": 1140, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1655, forks 318 (observed 2026-08-28T04:05:17.781342+00:00)

## What it is
PEGASUS is Google Research's implementation of transformer encoder-decoder models pre-trained with the Gap Sentences Generation objective for abstractive text summarization. It includes pre-trained checkpoints, a Flax-based PEGASUS-X variant, and training/evaluation code.

## Use cases
- summarize long documents into short abstracts
- fine-tune a pre-trained summarization model on my own dataset
- train a sequence-to-sequence model with gap sentence pre-training
- evaluate summarization models with ROUGE on benchmark datasets
- use PEGASUS-X for long-input summarization
- download pre-trained summarization checkpoints for research

## When to choose
- you need state-of-the-art abstractive summarization models with published benchmarks
- you want to reproduce or extend the PEGASUS/PEGASUS-X research
- you need pre-trained checkpoints for fine-tuning on custom summarization data

## When to avoid
- you need a production summarization API rather than research code
- you want a maintained library with frequent updates - the repo is largely in maintenance mode
- you need general-purpose text generation beyond summarization

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, transformers, llm-training
- domain: machine-learning, deep-learning
- platform: python, cloud
- tags: abstractive-summarization, sequence-to-sequence, pretraining, flax, tensorflow, research-models, text-summarization, natural-language-processing, gpu

## Member repositories
- google-research/pegasus (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:17.781342+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 2026-08-30T03:44:52.556259+00:00, confidence not recorded.
  - readme: https://github.com/google-research/pegasus (fetched 2026-08-28T04:05:17.781342+00:00, sha 5233eec63f81)
- Data as of 2026-08-30T08:39:29.467469+00:00.