This Paper proposed the prediction method of fuel consumption from vehicle information through OBD-II. We assumed RPM, TPS had a relationship with fuel consumption. We got the output as fuel-consumption from a vehicle RPM, TPS as input by using polynomial equation. We had modeling as quadric function with OBD-II data and fuel consumption data supported by automotive company in real. In order to verify the effectiveness of proposed method, 5 km real road-test was performed. The results showed that the proposed method can estimate precisely the fuel consumption from vehicle multi-data. It was observed that the proposed models using instantaneous engine RPM and TPS can predict the fuel consumption quite well with correlation coefficient were 76% and 88% respectively.