Nvidia designed the CUDA parallel-computing platform that helped make it a leading chipmaker by market value.
Nvidia introduced CUDA in 2006 as a platform and programming model for using graphics-processing units, or GPUs, for general-purpose computation. Instead of limiting GPUs to rendering images, CUDA enabled developers to use their many parallel processing cores for scientific computing, artificial intelligence, and other demanding workloads.
The company was founded in 1993 by Jensen Huang, Chris Malachowsky, and Curtis Priem. It first became widely known for graphics hardware used in personal computers and video games. Later, its GPUs became central to data centers and machine-learning systems, especially as neural-network training required large amounts of parallel computation.
CUDA is a software ecosystem rather than a chip. That distinction matters because Nvidia’s market position reflects both its hardware and the developer tools, libraries, and programming support built around its processors. AMD and Intel also make competing accelerators, but CUDA is uniquely associated with Nvidia.