INTRODUCTION: The global clinical and economic burden of type 2 diabetes is substantial. New GLP-1 receptor agonists have shown a multifactorial clinical profile with the potential to address many of clinical needs. OBJECTIVES: The objective of this study is to evaluate the cost-effectiveness of once-daily liraglutide vs. once-weekly exenatide in patients who had failed in metformin, sulfonylurea, or both treatments. METHODS: A Markov model is made to predict life expectance and QALYs of liraglutide and exenatide. Baseline characteristics are consistent with DURATION-6 clinical trial. Simulations were run over 35 years (one year as a cycle) from a third-party payer perspective. Future costs and benefits are discounted at 3%. 5 health states were included in the model: “No complications”, “Microvasular complications”, “Macrovasular complications”, “both complications” and “Death”.  Data was extracted from previous studies, head-to-head clinical trial, U.S. consumer Price Index, U.K. Perspective Diabetes Survey, Action in Diabetes and Vascular Disease trials, Action to Control Cardiovasular Risk in Diabetes trials and National Health Interview Survey data. The transition probabilities in the model vary by the age and gender of the patients to simulate the natural progression of type 2 diabetes.  RESULTS: Liraglutide is associated with improvement of 0.15 QALY. Even though it costs more than exenatide, it is still more cost-effective than exenatide. The incremental cost-effectiveness ratios per QALY gained with liraglutide is $138,282 (2013 US$), which is less than 3 times GDP per capita in 2013. Sensitivity analysis was done. Figures in the model were adjusted reasonably, and the results remain robust. In other word, liraglutide is more cost-effective than exenatide. CONCLUSIONS: Long-term projections indicated that liraglutide (injected daily) is more cost-effective than exenatide (applied weekly).