Tuberculosis, which is caused by single infectious agent Mycobacterium tuberculosis, is one of the most important infectious diseases. Tuberculosis is a major public health problem with approximately 2 million annual deaths. In the present study QSAR analysis of a series of substituted 2-polyfluoroalkyl and 2-Nitrobenzylsulphanyl benzimidazoles was performed using V-LIFE MDS 3.0 software 2D QSAR models were developed using partial least square (PLS) and variable selection methods. Out of 10 models developed. the two best 2D QSAR models having highest correlation coefficient and cross validated squared correlation coefficient were selected for further study, which were r2 = 0.9013, q2 = 0.7676, F test = 73.0375 pred_r2 = - 0.1772, pred_r2se = 0.5646 and r2 = 0.8441, q2 = 0.7088, F test = 27.0819 pred_r2 = -0.1082, pred_r2se = 0.5478 . Two 3D QSAR models were developed using KNN-MFA method, combined with simulated annealing selection procedure. Out of two models developed the best 3D QSAR model having highest cross validated squared correlation coefficient was selected for further study, which is q2 = 0.6765 pred_r2se = 0.5312 and q2 = 0.7747, pred_r2se = 0.8455 . A quantitative structure activity relationship study on a series of Halogenbenzimidazoles analogues was made using combination of various thermodynamic electronic and spatial descriptors. Several statistical expressions were developed using stepwise multiple liner regression analysis. The best quantitative structure activity relationship models were further validated by leave-oneout method of cross-validation.