Abstract: Stochastic hybrid systems are a class of stochastic models that incorporates both continuous- and discretestate dynamics. They have wide applications in modeling various processes, such as communication networks [4] and biochemical reactions [6]. In this work, we focus on continuous-time stochastic hybrid systems (CTSHS); specifically, we propose a framework using Metric Interval Temporal Logic (MITL) [1] to describe the systems’ behavior and verifying the MITL formulas via model reduction and sampling.