Nvidia created the CUDA parallel-computing platform.
CUDA, short for Compute Unified Device Architecture, lets developers use compatible Nvidia graphics-processing units for general-purpose computing as well as graphics. Nvidia introduced the platform in 2006, helping make its GPUs useful for scientific computing, engineering, artificial intelligence, and other workloads.
The technology became particularly important because GPUs can perform many calculations simultaneously. That design is well suited to matrix operations and neural-network training, which helped Nvidia become a central supplier for modern AI infrastructure.
CUDA is not the same thing as a graphics card or a programming language alone. It is a software and hardware computing platform built around Nvidia GPUs. Nvidia’s strong position in accelerated computing became a major factor in its rise among the world’s most valuable listed companies.