The main objective of the current study is to compute entropy generation impact on heat and mass transfer characteristics of magnetic nanofluid flow across the heated stretching sheet for high temperature differences. Most of studies are found to deal with a small temperature difference between the surface and ambient nanofluid. However, the circumstances arise where this temperature difference is high due to entropy generation. The primary focus of this study is to reduce excessive heating by using magnetic field as an insulating material. The density of the fluid is assumed as a function of temperature. The impact of thermophoresis and Brownian motion is also used along the stretching sheet to compute heat transfer characteristics. The coupled partial differential equations are converted into ordinary differential equations by using stream functions and similarity transformation. The numerical and graphical results are obtained for various governing parameters by using Keller box simulation. The velocity, temperature and concentration distribution are examined for each reducing parameter such as density parameter, Eckert number, Prandtl number, thermophoresis number, Brownian motion parameter, magnetic force parameter and buoyancy parameter. The main finding is that velocity and temperature distribution improves for lower density parameter. It was found that heat and mass transfer rate enhances as Brownian motion improves. Moreover, the skin friction and heat transfer rate increases as entropy generation increases.