provides a big jump in AI performance by introducing two-dimensional matrix registers.
Advanced Vector Extensions (AVX)
a set of instructions to perform floating-point operations on multiple data elements in parallel.
SSE
an extension to perform single instruction, multiple data (SIMD) operations on multiple data elements in parallel
BMI
Bit manipulation instructions.
F16C
an extension to the x86 instruction set architecture that provides support for converting between half-point precision and standard IEEE single-precision floating-point formats.
../Architecture