PyPI page
Home page
Author:
None
License:
Apache-2.0 AND MIT
Summary:
NVIDIA cuDNN Frontend — Python and C++ Graph API with SOTA attention (SDPA / Flash Attention), MoE grouped GEMM fusions, and FP8/MXFP8 kernels for Hopper and Blackwell GPUs.
Latest version:
1.28.0
Required dependencies:
cuda-tile
Optional dependencies:
apache-tvm-ffi
|
cuda-python
|
nvidia-cutlass-dsl
Downloads last day:
153,225
Downloads last week:
825,424
Downloads last month:
3,954,106