nvcuda::wmma

E1943853 UNEXPLORED

nvcuda::wmma is a CUDA C++ namespace providing warp-level matrix multiply-accumulate primitives that leverage NVIDIA Tensor Cores for high-performance mixed-precision linear algebra.

Try in SPARQL Jump to: Surface forms Referenced by

All labels observed (1)

Label	Occurrences
nvcuda::wmma canonical	1

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

WMMA API → namespace → nvcuda::wmma ⓘ