Low cycle fatigue (LCF) and creep fatigue interaction (CFI) loadings are the main factors resulting in the failure of many critical components in the infrastructure of power plants and aeronautics. Accurate prediction of life spans under specified loading conditions is significant for the design and maintenance of components. In the present study, various LCF and CFI tests are conducted to investigate the effects of temperature, strain amplitude, hold time and hold direction on the fatigue life of P92 steel. To predict fatigue life under different experimental conditions, various conventional life prediction models are evaluated and discussed. Moreover, a new empirical life prediction model is proposed based on the conventional Manson-Coffin-Basquin (MCB) model. The newly proposed model is able to simultaneously consider the effects of temperature, strain amplitude, hold time and hold direction on predicted life. The main advantage is that only the known input experimental parameters are required to perform the prediction. In addition to the validation made through the experimental data of P92 steel conducted in the present paper, the model is also verified through numerous experimental data reported in the literature for various 9–12% Cr steels.