Abstract Objectives This study was designed to assess the cost-effectiveness of vildagliptin versus glimepiride as add-on to metformin in the management of type 2 diabetes mellitus (T2DM) patients in the Greek healthcare setting. Methods A cost-effectiveness model was designed, using MS Excel, to compare two treatment strategies. Strategy 1 consisted of first-line metformin, followed by metformin + vildagliptin in second-line, while strategy 2 consisted of first line metformin, followed by metformin + glimepiride in second line. Subsequent lines were the same in both strategies and consisted of metformin + basal insulin and metformin + basal + rapid insulin. Clinical data and utility decrements relating to diabetes complications were taken from the published literature. Only direct medical costs were included in the analysis (cost base year 2014), and consisted of drug, adverse events and comorbidity costs (taken from local officially published sources and the literature). The perspective adopted was that of the Social Insurance Fund. The time horizon was lifetime, and future costs and outcomes were discounted at 3.5% per annum. Results Adding vildagliptin to metformin increased drug costs compared with adding glimepiride to metformin (€2853 vs. €2427, respectively). However, this increase was offset by a decrease in the costs of associated comorbidities (€4393 vs. €4539) and adverse events (€2757 vs. €3111), resulting in a lower total cost of €74 in strategy 1 compared with strategy 2. Comorbidities were the largest cost component in both strategies, accounting for 43.9 and 45.0% in strategies 1 and 2, respectively. Strategy 1 was also associated with increased life-years (LYs, 0.11) and quality-adjusted life-years (QALYs, 0.11) compared with strategy 2. Strategy 1 is therefore dominant, as it is associated with both lower overall costs and increased effectiveness. Conclusions Vildagliptin as add-on treatment to metformin in the management of T2DM in Greece appears to be dominant versus. glimepiride in terms of both cost per LY and cost per QALY gained.