I'll just answer this for myself by testing. It seems to still use MSVCs qmath.h or similar native library implementation:
#include <AssembleIt.au3>
Local $double_qmathPow
Local $asmstrc = DllStructCreate("double __x;double __y;")
$asmstrc.__x = 5.8
$asmstrc.__y = 0.9535
Func _QMATH_NAKED__QMATH_INLINE_double__QMATH_LINK_qmathPow()
_(" use32 ")
_(" fld qword[esp+12] ")
_(" fld qword[esp+4] ")
_(" ftst ")
_(" fstsw ax ")
_(" sahf ")
_(" jz pow_zero ")
_(" fyl2x ")
_(" fld1 ")
_(" fld st1 ")
_(" fprem ")
_(" f2xm1 ")
_(" faddp st1, st0 ")
_(" fscale ")
_(" pow_zero: ")
_(" fstp st1 ")
_(" ret ")
EndFunc
;_AssembleIt($double_qmathPowurntype, $Name_of_Func_with_code, $Type1 = "type", $Param1 = 0, $Type2 = "type", $Param2 = 0....up to 20 params)
$double_qmathPow = _AssembleIt("double", "_QMATH_NAKED__QMATH_INLINE_double__QMATH_LINK_qmathPow", "double", $asmstrc.__x, "double", $asmstrc.__y) ;thats all^^
$double_autoit = $asmstrc.__x ^ $asmstrc.__y
ConsoleWrite(@LF)
ConsoleWrite("+> (FASM)qmathPow: " & $double_qmathPow & @LF)
ConsoleWrite("+> (AutoIt)^ : " & $double_autoit & @LF)
minxomat's post in Details on how AutoIt handles fractional exponents. was marked as the answer