Objective: Implantation of a ventricular assist device (VAD) is a seminal therapeutic option for patients with terminal cardiac failure. A growing number of VAD patients are successfully bridged to transplantation, or can even live permanently with the device. However, the success is restricted by frequent severe complications. Haemolysis is a relevant adverse effect of several VAD types, which is the result of destruction of red blood cells, reduced by wall shear stress, flow acceleration and interaction with artificial surfaces. The CH-VAD, a small implantable continuous-flow blood pump, featuring a magnetically levitated impeller and enough hydrodynamic performance, was under development and completed for a 60-days animal implantation experiment in 6 sheep. The goal of this study is to validate the hemolysis of the pump through in vitro and in vivo studies. Methods: A series of in vitro tests was quantified experimentally by using in vitro circulation loop system according to ASTM F1841, the standard practice for the assessment of hemolysis in continuous-flow blood pumps. The hemolysis test in vivo was performed during a 60-days ovine model implantation, which was being conducted under the Institutional Animal Care and Use Committee (IACUC) protocol 05-0600 1. Results in vitro tests showed that the average normalized index of hemolysis (NIH) value of the VAD was 0.007 mg/l. The hemolysis in vivo was evaluated based on the amount of free hemoglobin in the plasma, and which showed that the free hemoglobin level in plasma peaked at 0.95 mg/l on the fifth postoperative day and then returned to an acceptable range of 6.0 mg/dL. Conclusion The magnetic levitation left ventricular assist device has good hemolytic performance. These acceptable performance results supported proceeding initial clinical trail conditions.