The objective of this paper is to present our experience in the commissioning and quality assurance (QA) for intensity-modulated radiotherapy (IMRT) using the dynamic multileaf collimator (dMLC), sliding window technique. The connectivity and operation between all IMRT chain components were checked on the Varian equipment. Then the following tests were performed: stability of leaf positioning and leaf speed, sensitivity to treatment interruptions (acceleration and deceleration), evaluation of standard field patterns, stability of dMLC output, segmental dose accuracy check, average leaf transmission, dosimetric leaf separation, effects of lateral disequilibrium between adjacent leaves in dose profiles, and multiple carriage field verification. Standard patterns were generated for verification: uniform field, pyramid, hole, wedge, peaks, and chair. Weekly QA protocol includes sweeping gap output, garden fence test (narrow bands, 2 mm wide, of exposure spaced at 2-cm intervals), and segmental dose accuracy check. Monthly QA includes sweeping gap output at multiple gantry and collimator angles, sweeping gap off-axis output, picket fence test (eight consecutive movements of a 5-cm wide rectangular field spaced at 5-cm intervals), stability of leaf speed and leaf motor current test (PWM test). Our patient QA procedure consists of an absolute dose measurement for all treatment fields in the treatment condition, analysis of actual leaf position versus planned leaf position (dynalog files) for each treatment field, film relative dose determination for each field, film relative dose determination for the plan (all treatment fields) in two axial planes, and patient positioning verification with orthogonal films. The tests performed showed acceptable results. After over one year of IMRT treatment, the routine QA machine checks confirmed the precision and stability of the IMRT system.