Tensor Core
A Tensor Core is an NVIDIA execution unit for supported matrix multiply-accumulate operations. It computes small matrix products that software combines into larger operations.
How it works
A multiply-accumulate updates a tile of C using A × B. Instructions impose supported shapes, layouts, operand formats and accumulator formats. Libraries arrange tiles and data movement so these matrix operations receive their inputs fast enough.
Work through an example
For a dense M × K times K × N multiplication, the usual work convention is approximately 2MKN floating-point operations. Compare throughput with a peak for the same precision and density assumption.
Where the term can mislead
A published structured-sparse peak is not the dense peak. Tensor Core availability also does not guarantee that a particular operator selects a Tensor Core kernel.
Connect the idea
Take it into practice
Continue with the existing concept, apply it to an interview problem, or follow a course lesson. Access requirements are shown before you open a page.
New to GPUs? Start with the free execution-model concept.
Sources and scope
Examples here are teaching calculations, not measured GPU performance. Consult the linked documentation for the exact device, instruction and tool version. Source review: 25 October 2026.