In this paper, I have attempted the quantitative analysis and comparison of the Mongolian dialects and languages. Currently, the computational and statistical analytic methods are used widely in every field of the linguistics, and one of those methods are applied in this paper. 16 Mongolian dialects and languages are analysed by the N-gram language model. After the bigram vocabulary sets of the 16 Mongolian dialects and languages are extracted, the positive log probabilities of each dialect and language are calculated automatically. Then, 15 dialects and languages are compared with the Zhenglanqi dialect. The positive log probabilities are used in this comparison method. After the comparison work, every dialect and language are ordered in the distance from the Zhenglanqi dialect by the degrees of the phonotactic similarities.