Current maintenance, having a great impact on the safety, reliability and economics of a gas turbine, becomes the major obstacle for the application of gas turbines in energy field. An effective solution is to process condition based maintenance (CBM) thoroughly for gas turbines. Maintenance of high temperature blade, accounting for the most of the maintenance costs and time, is the crucial section of gas turbine maintenance. The suggested life of high temperature blade by original equipment manufacturer (OEM) is based on several certain operating conditions, which is used for time based maintenance (TBM). Thus, for the requirement of gas turbine CBM, a damage evaluation model is demanded to estimate the life consumption online. A physics-based model is built, consisting of thermodynamic performance simulation model, stress estimation model, thermal estimation model, and interactive damage analysis model. Unmeasured parameters are simulated by the thermodynamic performance simulation model, as the input of the stress estimation model and the thermal estimation model. Due to the ability to analyze online data, this model can be used to calculate online damage and support CBM decision. Then the stress and temperature distribution of blades will become as the input of the creep damage analysis model and the fatigue damage analysis model. The interactive damage of blades will be evaluated based on the creep and fatigue analysis results. To validate this physics-based model, it is used to calculate the lifes of high temperature blade under several certain operating conditions. And the results are compared to the suggestion value of OEM. An application case is designed to evaluate the application effect of this model. The result shows that the relative error of this model is less than 10.4% in selected cases. And it can cut overhaul costs and increase the availability of gas turbines significantly. Finally, a simple application of this model is proposed to show its functions. The physical-based damage evaluation model proposed in this paper is found to be a useful tool to tracing the online life consumption of a high temperature blade, to support the implementation of CBM for gas turbines, and to guarantee the reliability of gas turbines with lowest maintenance costs.