Simulation of nuclide inventory evolution in the reactor during the entire irradiation history is of significance to the design of fuel component and core layout scheme and nuclear waste disposal. To perform accurate burnup calculations for fast reactor systems such as accelerator-driven sub-critical reactors, a nuclide depletion code IMPC-Depletion has been developed. Depletion of reactor material is described by thousands of coupled first-order ordinary differential equations, which is the stiff ODEs system. For IMPC-Depletion, algorithms including Chebyshev Rational Approximation Method (CRAM) and Transmutation Trajectory Analysis method (TTA) are implemented to solve Bateman equations. The parameter-extended XML file format is used as standard input. Decay, constant flux and power modes are supported. To verify IMPC-Depletion code, decay of Ra228, fixed-flux irradiation of Zircaloy-4 and 2.4% UO2, fixed-power irradiation of 3.2% UO2 are tested. Results are basically consistent with those of ORIGEN2.1. In addition to independent nuclides inventory analysis, IMPC-Depletion can also be coupled with Monte Carlo neutron transport code to solve burnup problem. Here, simple MOX pin cell and OECD/NEA fast reactor burnup benchmarks are employed to verify the Out-coupling Monte Carlo burnup code system IMPC-Burnup2.1, which couples OpenMC and IMPC-Depletion. Result parameters such as keff, reactivity changes and nuclides inventory values are reliable when compared with other participants.