Calculate Probability of Membership from kmeans in R

Question

How can I calculate the probability of membership with R's kmeans output?

The output of kmeans is as follows:

k <- kmeans(iris[-5], 3)
str(k)
# List of 9
# $ cluster     : int [1:150] 2 3 3 3 2 2 2 2 3 3 ...
# $ centers     : num [1:3, 1:4] 6.31 5.18 4.74 2.9 3.62 ...
# ..- attr(*, "dimnames")=List of 2
# .. ..$ : chr [1:3] "1" "2" "3"
# .. ..$ : chr [1:4] "Sepal.Length" "Sepal.Width" "Petal.Length" "Petal.Width"
# $ totss       : num 681
# $ withinss    : num [1:3] 118.65 6.43 17.67
# $ tot.withinss: num 143
# $ betweenss   : num 539
# $ size        : int [1:3] 96 33 21
# $ iter        : int 2
# $ ifault      : int 0
# - attr(*, "class")= chr "kmeans"

Is this enough information to copy what we get from Mclust?

library(mclust)
m <- Mclust(iris[-5])
head(m$z)
#           [,1]         [,2]
# [1,] 1.0000000 2.513157e-11
# [2,] 0.9999999 5.556411e-08
# [3,] 1.0000000 3.635438e-09
# [4,] 0.9999999 8.611811e-08
# [5,] 1.0000000 8.504494e-12
# [6,] 1.0000000 1.400364e-12

Obvious question is "Why not use mclust?". My data is too large to computationally do hierarchical clustering. I have tried with Mclust, NbClust, vegan, and many others. The call to dist that all of the functions use max out after a few hundred thousand rows.

I have seen some talk about probabilistic-D, and "soft" clustering, but I do not know how to implement it without changing the output of the original clusters from kmeans.

Edit I know that SAS is able to export probabilities with PROC FASTCLUS, but from what I hear it is taking a sample of the data to get the probabilities. That might be one route to take if I could figure out how it's doing the subsetting.