Nvidia, one of the world's largest companies by market capitalization, introduced its CUDA parallel-computing platform in 2006.
CUDA is a parallel-computing platform and application programming interface created by Nvidia. It allows software developers to use Nvidia graphics-processing units for general-purpose computing, rather than limiting them to graphics and image rendering. Nvidia announced CUDA in 2006, initially presenting it as a way to make GPU computing more accessible to programmers.
The platform helped popularize the use of GPUs for workloads involving many simultaneous calculations. Scientific modeling, engineering, financial analysis, image processing, and machine learning all benefited from this approach. CUDA programs commonly use the C and C++ programming languages, with additional support developed for other programming environments.
CUDA is sometimes confused with a graphics card or a GPU architecture. It is neither: it is a software platform and programming model that works with compatible Nvidia GPUs. Its growth became especially important as artificial-intelligence researchers adopted GPUs for training and running neural networks.