Flamelet Generated Manifold (FGM) has proven to be an efficient approach to model turbulent combustion across different regimes of combustion. The manifolds are generally created by solving laminar premixed or opposed flow configurations. Gas turbine combustors often involve many strong non-adiabatic events such as multiple temperature boundaries, quenching from cooling and effusion holes, conjugate heat transfer, soot radiation interaction, phase change from spray and the modulation of inlet conditions. The adiabatic assumption of the underlying flamelet generation in the FGM is, therefore, prone to errors in the prediction of flame speed, liner temperatures, and pollutant formation. In this work, a novel approach to generate fully non-adiabatic manifold is proposed and validated. The FGM manifold is created using a series of non-adiabatic flamelets, each flamelet is solved in one-dimensional physical space. The non-adiabatic flamelets are generated with an optimal combination of freely propagating and burner stabilized flames. This hybrid method of the flamelet configuration allows modeling large heat gain and loss without encountering any unrealistic temperature in the flamelet solution. Such fully non-adiabatic flamelets are then convoluted to generate a five-dimensional Non-adiabatic Flamelet Generated Manifold (NFGM) Probability Density Function (PDF.). The average properties such as temperature, mixture density, species concentration, rate of reaction, etc. from PDF are then coupled with the CFD solution. The non-adiabatic flamelets and corresponding NFGM is implemented into ANSYS Fluent software version 2020R1. This approach is validated first for canonical cases, followed by gas turbine like conditions of swirl stabilized methane fueled turbulent flame, developed at DLR Stuttgart as the PRECCINSTA combustor. The experimental data for this combustor is available for multiple operating conditions. A stable operating point (φ = 0.83, P = 30 kW) is chosen. The proposed nonadiabatic NFGM is used with Stress blended eddy simulation (SBES). The current NFGM-SBES results are compared with experimental data as well as the previously published works. The impact of modeling heat release in flamelet is used to analyze the M-shape versus V-shape flame transition and the peaks of the carbon monoxide in mixing shear layers. The findings from the current work, in terms of accuracy, validity and best practices while modeling NFGM-SBES are discussed and summarized. The improved results of NFGM compared to adiabatic FGM are encouraging and provides a potential high-fidelity tool for accurate, yet efficient modeling of turbulent combustion inside gas turbines.