The size of the melt pool directly regulates the mechanical properties, surface morphology, and dimensional accuracy in metal laser powder bed fusion (LPBF). Analytical modeling is a good choice for the simulation of metal LPBF as it provides a practical estimate of melt pool size, i.e., width and depth for any material with different process parameters. In the literature, several analytical models were described for melt pool size prediction. The prediction accuracy of such analytical models is poor at various linear energy density values, i.e., laser power divided by scan velocity. As a result, model calibration or a single-track characterization is essential for such analytical models. In this study, the Liu and Kannatey-Asibu (Trans ASME 115(2):34–41, 1993) model originally proposed for laser welding is applied for the melt pool width and depth prediction in metal LPBF. The MATLAB programs are developed to calculate temperature distribution during a single-track scan with several combinations of laser power and scan velocity. The melt pool sizes are estimated by matching predicted temperatures with liquidus temperature of alloy material. The predictions are made for Inconel625 and Ti-6Al-4V materials. The estimated values are validated by predicted values of the Eager and Tsai model and published experimental work in the literature. A good resemblance is observed between predicted and experimentally measured melt pool sizes. This model has shown superior accuracy in the prediction of melt pool size over the Eager and Tsai model. This model can be used for fast-track process parameter optimization for the printing of dense and defect-free parts.