facebookresearch/fairseq

Health v2 · maintenance only

Flags: archived

Adoption not part of the score

32231 stars · 6676 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

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

library · maturity maintenance

machine-learning deep-learning llm-training nlp speech-recognition machine-learning deep-learning artificial-intelligence speech-processing python pytorch sequence-modeling neural-machine-translation transformers research facebook-ai-research natural-language-processing linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="facebookresearch/fairseq")