Nvidia developed the CUDA parallel-computing platform, which became central to its modern AI-chip business.
Nvidia introduced CUDA in 2006 as a software platform and programming model for using graphics processing units for general-purpose computation. It allowed developers to run suitable workloads on Nvidia GPUs rather than relying only on central processing units.
The platform became important in scientific computing, machine learning, and artificial intelligence. Modern AI systems require enormous amounts of parallel calculation, and Nvidia’s combination of GPUs, networking, libraries, and development tools made its hardware especially influential in data centers.
Nvidia was founded in 1993 by Jensen Huang, Chris Malachowsky, and Curtis Priem. It originally became famous for graphics processors used in video games, but demand for AI infrastructure later transformed how investors valued the company. CUDA is Nvidia-specific; AMD offers ROCm, while Intel has its own software ecosystem, so these platforms are not interchangeable names.