Skip to content

Files

Latest commit

 

History

History
 
 

fbgemm_gpu

FBGEMM_GPU

FBGEMM_GPU CI FBGEMM_GPU-CPU Nightly Build FBGEMM_GPU-CUDA Nightly Build

FBGEMM_GPU (FBGEMM GPU Kernels Library) is a collection of high-performance PyTorch GPU operator libraries for training and inference. The library provides efficient table batched embedding bag, data layout transformation, and quantization supports.

FBGEMM_GPU is currently tested with cuda 12.1.0 and 11.8 in CI, and with PyTorch packages (2.1+) that are built against those CUDA versions.

Only Intel/AMD CPUs with AVX2 extensions are currently supported.

See our Documentation for more information.

Installation

The full installation instructions for the CUDA, ROCm, and CPU-only variants of FBGEMM_GPU can be found here. In addition, instructions for running example tests and benchmarks can be found here.

Build Instructions

This section is intended for FBGEMM_GPU developers only. The full build instructions for the CUDA, ROCm, and CPU-only variants of FBGEMM_GPU can be found here.

Join the FBGEMM_GPU Community

For questions, support, news updates, or feature requests, please feel free to:

For contributions, please see the CONTRIBUTING file for ways to help out.

License

FBGEMM_GPU is BSD licensed, as found in the LICENSE file.