Models of photovoltaic (PV) are significant in the design, study and control of renewable energy system. In these models, equivalent diode circuit model is widely researched and used because of its high precision. In diode circuit model, usually there are several parameters to be determined. To improve the performance of PV models, it is important to extract their unknown parameters exactly and quickly. However, because of the nonlinear of the V-I characteristic output of PV, it is difficult to obtain the parameters accurately. In this paper, a self-adaptive firefly algorithm is proposed to extract the parameters of single diode circuit model. Through introducing an adaptive mutation factor into the evolution process of a firefly algorithm, it improves the precision and stability of the solution. The proposed adaptive firefly algorithm is used to extract the parameters of single diode model which have 5 unknown parameters. By using measure output voltage and current data of PV, the proposed algorithm can extract parameters exactly and quickly. The proposed algorithm is compared with classic firefly algorithms and other algorithms in the paper. The results show that the effectiveness of the proposed algorithm.