fairseq2 and fairseq

Fairseq2 is the modernized successor to the original fairseq, designed to replace it with improved architecture and performance while maintaining conceptual compatibility with its predecessor's sequence-to-sequence modeling framework.

fairseq2
74
Verified
fairseq
53
Established
Maintenance 17/25
Adoption 11/25
Maturity 25/25
Community 21/25
Maintenance 2/25
Adoption 10/25
Maturity 16/25
Community 25/25
Stars: 1,122
Forks: 135
Downloads:
Commits (30d): 9
Language: Python
License: MIT
Stars: 32,190
Forks: 6,676
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No risk flags
Stale 6m No Package No Dependents

About fairseq2

facebookresearch/fairseq2

FAIR Sequence Modeling Toolkit 2

This toolkit helps AI researchers train and fine-tune custom AI models for various content generation tasks, such as creating new text, speech, or even translating between languages. You feed it large datasets of text, audio, or other sequences, and it outputs a trained AI model ready for deployment. This is for researchers specializing in natural language processing, speech technology, or other generative AI fields.

AI-research natural-language-processing speech-recognition machine-translation generative-AI

About fairseq

facebookresearch/fairseq

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

This toolkit helps researchers and developers build custom models for processing and generating text and speech. You can feed it raw text or audio, and it produces translated content, summarized documents, or other forms of generated language. It's designed for machine learning scientists, AI researchers, and engineers working on advanced language and speech applications.

natural-language-processing machine-translation speech-recognition text-generation AI-research

Scores updated daily from GitHub, PyPI, and npm data. How scores work