Load balancing is a challenging aspect on the increase in the performance and improvement of the utilization in the grid environment. Prediction of future load is a prime factor in balancing of load. Many attempts were made in the past on prediction of load with simple classifier approaches. There are many mathematical and machine learning models available on the numeric predictions of the parameters on the available data. The objective is to anticipate the demand of the load in the grid using numeric prediction methods with machine learning models. In this research, linear regression, Gaussian process and generalized linear model are suggested among many mathematical and machine learning models on the prediction of the load against the standard data set with respect to the suitability and close association of the parameters of load. A comparison is made on the performance of the prediction of the load among these selected models.