NVIDIA CUDA 13.3 Enhances GPU Development with Tile Programming in C++, Compiler Autotuning, and Python Updates
NVIDIA CUDA 13.3 introduces tile-based C++ kernel programming with automatic low-level GPU management
“NVIDIA CUDA Tile programming in C++, enables high-level, tile-based kernel development that automatically manages complex low-level GPU details for optimal performance and portability.”
NVIDIA released CUDA 13.3, featuring Tile programming in C++ that abstracts low-level GPU memory and compute details for developers. The update also includes compiler autotuning and Python improvements, targeting performance portability across GPU architectures including Compute Capability 9.0. While significant for GPU software developers, this is an incremental tooling release rather than a foundational AI industry shift.