Explore projects

Ongoing research training transformer language models at scale, including: BERT & GPT-2

physics meets neural networks

Mamba SSM architecture

LLM inference in C/C++

an easy-to-use knn-mt toolkit

Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

Fast Hadamard transform in CUDA, with a PyTorch interface

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