CUDA is NVIDIA's software platform that lets programmers use NVIDIA GPUs for general computing, including training and running AI models.
In plain words
GPUs began as chips for 3D graphics. CUDA, introduced in 2006, let any program use a GPU's power without going through graphics tools. A GPU runs thousands of small tasks at once, while a CPU runs a few very fast; AI maths suits the GPU. The CUDA Toolkit gives developers the tools to build GPU programs, and NVIDIA's CUDA-X libraries, built on CUDA, speed up AI and other fields.
A quick example
Imagine training a model on an NVIDIA GPU with a popular AI framework. You may never write CUDA yourself, but the heavy maths runs through CUDA-based libraries that spread the work across thousands of GPU threads at once.
Where you'll meet it
- Deep learning frameworks running on NVIDIA GPUs.
- Scientific simulation and data analytics.
Related
Sources
- DocsCUDA Platform for Accelerated Computing, NVIDIA Developer · read 29 Sept 2026
- DocsCUDA Programming Guide: Introduction, NVIDIA · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.