Antibiotic-resistant Klebsiella pneumoniae has emerged as a critical public health threat worldwide. Understanding the antimicrobial resistance mechanisms of multidrug-resistant K. pneumoniae (MDR-Kp) and its prevalence in time and space would provide clinical significance for managing pathogen infection. Eighteen clinical MDR-Kp strains were analyzed by whole genome sequencing (WGS), and the antimicrobial resistance genes and associated resistance mechanisms were compared with results obtained from the conventional microbiological test (CMT). The sequence homology across strains in our study and those previously collected over time from a wide geographical region was assessed by phylogenetic analysis. MDR-Kp strains were collected from eighteen patients who had received empirical treatment before strain collection, with sputum (83.3%, 15/18) being the primary source of clinical samples. The commonly received treatments include β-lactamase inhibitors (55.6%, 10/18) and carbapenems (50%, 9/18). Using CMT, we found that all 18 strains were resistant to aztreonam and ciprofloxacin, while 14 (77.8%) showed resistance to carbapenem. Polymyxin B and tigecycline were the only antibiotics to which MDR-Kp strains were sensitive. A total of 42 antimicrobial resistance mechanisms were identified by WGS, surpassing the 40 detected by the conventional method, with 25 mechanisms shared between the two techniques. Despite a 100% accuracy rate of WGS in detecting penicillin-resistant strains, the accuracy in detecting cephalosporin-resistant strains was only at 60%. Among all resistance genes identified by WGS, Klebsiella pneumoniae carbapenemase-2 (KPC-2) was present in all 14 carbapenem-resistant strains. Phenotypic analysis indicated that sequence type (ST) 11 isolates were the primary cause of these MDR-Kp infections. Additionally, phylogenic clustering analysis, encompassing both the clinical and MDR-Kp strains previously reported in China, revealed four distinct subgroups. No significant difference was observed in the sequence homology between K. pneumoniae strains in our study and those previously collected in East China over time. The application of WGS in identifying potential antimicrobial-resistant genes of MDR-Kp has demonstrated promising clinical significance. Comprehensive genomic information revealed by WGS holds the promise of guiding treatment decisions, enabling surveillance, and serving as a crucial asset in understanding antibiotic resistance.