This is a super duper fast implementation of the kmeans clustering algorithm. The code is fully vectorized and extremely succinct. It is much much faster than the Matlab builtin kmeans function. The kmeans++ seeding algorithm is also included (kseeds.m) for good initialization. Therefore, this package is not only for coolness, it is indeed practical.
Please try the demo script in the package.

Detail explanation of this algorithm can be found in following blog post:
http://statinfer.wordpress.com/2011/12/12/efficient-matlab-ii-kmeans-clustering-algorithm/

This function is now a part of the PRML toolbox (http://www.mathworks.com/matlabcentral/fileexchange/55826-pattern-recognition-and-machine-learning-toolbox).