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 and 11.8 in CI, and with PyTorch packages (2.1+) that are built against those CUDA versions.
See the full Documentation for more information on building, installing, and developing with FBGEMM_GPU, as well as the most up-to-date support matrix for this library.
For questions, support, news updates, or feature requests, please feel free to:
- File a ticket in GitHub Issues
- Post a discussion in GitHub Discussions
- Reach out to us on the
#fbgemmchannel in PyTorch Slack
For contributions, please see the CONTRIBUTING file for
ways to help out.
FBGEMM_GPU is BSD licensed, as found in the LICENSE file.