# facebookresearch/fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

Repository: https://github.com/facebookresearch/fairseq
Canonical: https://ross.abutalabs.com/products/fairseq
Language: Python
License: MIT
License Family: permissive
Topics: python, pytorch, artificial-intelligence
Archived: true
Last push: 2025-09-30T11:16:06+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 44, release rhythm 8, longevity 100
- inputs: {"age_days": 3291, "days_push": 337, "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 32231, forks 6676 (observed 2026-08-28T04:11:57.475264+00:00)

## What it is
Fairseq is a PyTorch-based sequence modeling toolkit from Facebook AI Research for training custom models for translation, summarization, language modeling, and other text generation tasks. It ships reference implementations of many influential sequence modeling papers, including Transformers, CNN seq2seq, and wav2vec speech models.

## Use cases
- train a neural machine translation model
- train custom language models in pytorch
- run speech recognition with wav2vec
- reproduce sequence-to-sequence research papers
- fine-tune transformer models for text generation
- build summarization models

## When to choose
- you need well-tested reference implementations of classic seq2seq and NMT papers
- you want a research-grade PyTorch toolkit for translation or language modeling
- you need wav2vec-style speech pretraining models

## When to avoid
- you want the latest LLM fine-tuning ecosystem rather than classic sequence modeling
- you need a production inference server rather than a training toolkit
- you prefer newer frameworks like Hugging Face Transformers for general NLP

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, llm-training, nlp, speech-recognition
- domain: machine-learning, deep-learning, artificial-intelligence, speech-processing
- platform: python
- tags: pytorch, sequence-modeling, neural-machine-translation, transformers, research, facebook-ai-research, natural-language-processing, linux, macos, gpu

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:57.475264+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 2026-08-29T16:52:11.189629+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/fairseq (fetched 2026-08-28T04:11:57.475264+00:00, sha 0ba269ba90a3)
- Data as of 2026-08-30T08:39:29.467469+00:00.