Özer Cem KELAHMETKELAHMET
Glossary · MLOps and AI infrastructure

What is CUDA? Definition and meaning

A programming model and toolkit for running general-purpose code on NVIDIA GPUs.

Definition

CUDA: A programming model and toolkit for running general-purpose code on NVIDIA GPUs.

In short: NVIDIA's GPU computing platform.

When it is used

GPU kernels and the base of deep-learning libraries.

Where it fits

CUDA belongs to Hardware in the MLOps and AI infrastructure topic. These are the other terms in the same group:

  • NCCL: A library that implements fast all-reduce and other communication between GPUs.
  • Cloud TPU: Google's custom accelerators for machine-learning workloads.
  • NVIDIA H100: A GPU with Transformer Engine and fast interconnect, widely used for training and inference.
  • NVIDIA H200: A Hopper GPU with 141 GB of HBM3e memory.
  • NVIDIA B200: A GPU with a new architecture, larger memory and support for lower-precision formats.

Learn more

The full guide to CUDA explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.

Previous term: Cross-validation · Next term: Curriculum learning

Last updated Nov 18, 2026
Type to search.
    ↑↓ navigate↵ open