After clicking “Watch Now” you will be prompted to login or join.


Click “Watch Now” to login or join the NVIDIA Developer Program.


CuPy Overview: NumPy Syntax Computation with Advanced CUDA Features

Crissman Loomis, Preferred Networks | Emilio Castillo, Preferred Networks

GTC 2020

We'll introduce CuPy, describing the advantages of the library and how it is cleanly exposing in Python multiple CUDA state-of-the art libraries such as cuTENSOR or cuDNN. Discover the CuPy advantages and how they can use it to experience performance gains in their NumPy codes without any major changes.

View More GTC 2020 Content