Virtual power plants (VPPs) can participate in power system dispatch and provide considerable regulation capability, thereby supporting renewable energy integration and secure grid operation. However, due to multiple sources of uncertainty within VPPs, their available regulation capability is highly uncertain, which may result in insufficient support when a VPP is called upon by the grid. To address this issue, a stochastic power system dispatch method considering credible regulation capability of VPPs is proposed. First, the definition, quantitative indicator, and evaluation process of VPP credible regulation capability are presented. Then, considering the uncertainty of regulation capability that VPPs provide and the stochastic nature of wind power output, chance constraints on power system reserve requirements are constructed, and a power system stochastic dispatch model with embedded chance constraints is established. Furthermore, to reduce the computational complexity of the model, a deterministic transformation method suitable for the system reserve chance constraints is proposed. Finally, case studies show that the proposed method effectively reduces system operational risk, improves the system’s ability to cope with renewable generation uncertainty, and promotes secure power system operation.