Laser powder bed fusion (LPBF) of Zr-based bulk metallic glasses (BMGs) has recently attracted attention due to its capacity for microstructure control, the potential for custom-tailored properties, and its versatile applicability across various industries. It is crucial to strike a balance between relative density, relaxation, and crystallinity tailored to a specific application when 3D printing amorphous components. However, prior findings have not revealed an exclusive study concerning low fractions of crystallinity (<5 vol%) and atomic-scale effects during the LPBF process, e.g. the relaxation degree of the amorphous structure. This study employs a systematic experimental approach, complemented by semi-analytical modeling, to comprehensively recognize the nano- and macro-scale amorphous structure and the associated mechanical behavior. Differential scanning calorimetry (DSC) along with flexural stress-strain measurements disclose that the highest relaxation and crystallization enthalpies, which are obtained for samples printed with lower heat input, are not in correlation with the desired mechanical properties. On the contrary, samples printed with a heat input of ΔH=44 reveal the maximum density range (up to 99.97 %) and showcase laboratory-XRD amorphous structure. This results in the highest flexural strength (up to 2080 MPa) and elastic deformation range (up to 2.6 %), along with the lowest Young's modulus span (60–70 GPa). In this case, the crystallinity (0.02–0.2 vol%) associated with the desired mechanical properties is calculated from the modeling results. The outcomes indicate increased heating/ cooling rates corresponding to the amplified heat input, with cooling rates reaching up to 5×104 °C/s and heating rates up to 6×105 °C/s. However, underlying layers are subjected to reduced heating and cooling rates. This observation not only validates the intricate thermal dynamics in LPBF but also elucidates the crystalline formation in heat-affected zones (HAZs) with associated lower heating and cooling rates.