This work focuses on the problem of finite-time multiple delay-dependent filter-based fault detection (FD) and fault-tolerant control (FTC) for semi-Markovian jump random nonlinear systems subject to state and input constraints. There are model uncertainties, multiple time-varying delays, nonlinear dynamics as well as faults in these systems. This study is the first time to try. In addition, this article both considers stochastic finite-time boundedness and input–output finite-time mean square stabilization. A semi-Markovian jump filter, a semi-Markovian jump controller and a fault weighting system are developed to achieve FD and FTC. Then, an augmented closed-loop nonlinear system can be established. Moreover, multiple delay-dependent sufficient conditions are produced by means of stochastic Lyapunov function in this framework of linear matrix inequalities. Finally, an example is shown to bring out the effectiveness of the method addressed in this study.