There is RCPSS — Compute Reciprocal of Scalar Single-Precision Floating-Point Values (but there is no RCPSD — Compute Reciprocal of Scalar Double-Precision Floating-Point Values).

But there is VRCP28SD (Approximation to the Reciprocal of Scalar Double-Precision Floating-Point Value with Less Than 2^-28 Relative Error) for AVX-512