NVIDIA CUDA 13.3 Simplifies GPU Development with Tile Programming in C++
NVIDIA released CUDA 13.3 with Tile programming in C++ — developers can now write GPU kernels at a high level of abstraction without manual optimization. Added

◐ Listen to article
NVIDIA released CUDA 13.3 with Tile programming in C++ — developers can now write GPU kernels at a high level of abstraction without manual optimization. Added automatic compiler autotuning and improved Python support. Works on new GPU generations (Compute Capability 9.0 and higher).