Unmanned aerial vehicles (UAVs) are already widely used to provide both relay services and enhanced information coverage to the terrestrial Internet of Things (IoT) networks. IoT devices may not be able to handle heavy computing tasks due to their severely limited processing capability. In this paper, a multi-UAV deployment for mobile edge computing (MEC) enhanced IoT architecture is designed, where multiple UAVs are endowed with computing offloading services for ground IoT devices with limited local processing capabilities. In order to balance the load of UAVs, this paper proposes a multi-UAV deployment mechanism which is based on the difference evolution (DE) algorithm. Meanwhile, the access problem for IoT nodes is formulated as a generalized assignment problem (GAP), and then an approximate optimal solution scheme is used to solve the problem. Based on this, we realize the load balancing of multiple UAVs, guarantee the constraint of coverage range and meet the quality of service (QoS) of MEC networks. Finally, sufficient simulations prove the effectiveness of our proposed multi-UAV deployment algorithm.