numpy random usage validity

I would like to generate 1000 gaussian random numbers.

Is those 2 equivalents in term of mathematical/stat results ? (not speed, but stat results).

direct

import numpy as np
np.random.seed(123)
x= np.random.normal(10, 1000)

With Loop:

 import numpy as np
 np.random.seed(123)
 for(i  in range(0,1000)):
   x[:,i]= np.random.rand(10, 1)

Reason is to paralell sampling... I looked into other questions, there is no indication about the Mersenne Twister implementation in numpy.