GPU
GPU (noun) means a chip designed to perform very large numbers of simple calculations at the same time, originally to draw images on a screen and now also used for tasks such as video editing, scientific computing and training machine-learning models. Example: “Rendering the scene took twenty minutes on the CPU and under a minute on the GPU.”
How to Use GPU
Learner’s notesIn plain EnglishA chip that does thousands of simple sums at once, used for graphics and AI.
Standard technical vocabulary; also common among gamers, where graphics card is used interchangeably for the physical card.
Say each letter separately. In shopping contexts GPU usually means the whole card, not just the chip.
Trace the full origin ↓Fill the Gap
Can you complete this real example?
Rendering the scene took twenty minutes on the CPU and under a minute on the _____.
Etymology
An initialism of graphics processing unit.
Synonyms
View more →Frequently Asked Questions
What is the difference between a GPU and a CPU?
A CPU has a few powerful cores that handle varied, unpredictable work one step after another, which suits operating systems and general software. A GPU has thousands of far simpler cores that all run the same operation across many pieces of data at once. Neither is an upgraded version of the other.
What is the difference between integrated and discrete graphics?
Integrated graphics is a GPU built into the same package as the CPU and sharing the computer's main memory. A discrete GPU is a separate chip on its own card with its own dedicated memory, so it generally handles heavier graphics work.
Why are GPUs used for artificial intelligence?
Training and running neural networks mostly comes down to multiplying very large grids of numbers, which is exactly the kind of repetitive parallel arithmetic a GPU is built for. That is why the same hardware sold for games became central to machine learning.