Introduction: For a long time, the Mw\Pharm software suite (MEDI\WARE, Prague, Czech Republic / Groningen, Netherlands) has been used for PK/PD modelling in therapeutic drug monitoring (TDM). Objective: The aim of this study was to find the best model in the newer Windows version of Mw\Pharm++ 1.3.5.558 (WIN) for patient on intermittent hemodialysis. Methods: Twenty-two adult patients with mean age 65±15 years, body weight 88±25 kg, were repeatedly examined for vancomycin. Cmin and concentration before hemodialysis (CBH) predicted by Windows models “vancomycin_dialysis_c2 “ (WINd), ”vancomycin_adult_C2“ (WINa), and DOS 3.30 models ” vancomycin (dialysis)“ (DOSd), ”vancomycin adult“ (DOSa) were compared with the with the measured value and with DOSd model. Statistics: Percentage prediction error (%PE) calculated as (predicted–measured)/measured or (predicted-DOSd)/DOSd, RMSE, Blandt-Altman bias, Pearson’s coefficient of rank correlation (R), Student’s t-test. Statistical analysis was performed using GraphPad Prism version 5.00 for Windows. Results: Models WINd and DOSd produced better Cmin prediction then respective adult models. DOSd model produced lower %PE, BlandtAltman bias and Pearson R, while slightly higher RMSE than WINd model. CBH prediction (available only in 7 cases) was better by WINa compared to WINd, while similar by DOSa and DOSd (Table 1). WINd and DOSd models use higher population V1 and Clm while lower fr than WINa and DOSa models (Table 2). V1, k12 were higher, while k21 was lower in all models than respective population data. Clm was higher in DOSa, while lower in WINa and fixed in WINd. Fr was lower in WINa and DOSa. Conclusion: DOS and WIN models are not identical despite similar name and the same population data. Dialysed models should be use for TDM in dialysed patients. DOSd model produced the best prediction while WINd was poorer. Implementation of DOSd model into WIN version of MW\Pharm is recommended.