Comprehending water dynamics is crucial in various fields such as water desalination, ion separation, electrocatalysis, and biochemical processes. While ab-initio molecular dynamics (AIMD) accurately portray water’s structure, computing its dynamic properties over nanosecond timescales proves cost-prohibitive. This study employs machine learning potentials (MLPs) to accurately determine the dynamical properties of liquid water with ab-initio accuracy. Our findings reveal diversity in the calculated diffusion coefficient (D) and viscosity of water (η) across different methodologies. Specifically, while the GGA, meta-GGA, and hybrid functional methods struggle to predict dynamic properties under ambient conditions, whereas methods on the higher level of Jacob’s ladder of DFT approximation perform significantly better. Intriguingly, we discovered that all D and η adhere to the established Stokes-Einstein (SE) relation for all the ab-initio water. The diversity observed among different methods can be attributed to distinct structural entropy, affirming the applicability of excess entropy scaling relations across all functionals. The correlation between D and η provides valuable insights for identifying the ideal temperature to accurately replicate liquid water’s dynamic properties. Furthermore, our findings can validate the rationale behind employing artificially high temperatures in the simulation of water via AIMD. These outcomes not only pave the path toward designing better functionals for water but also underscore the significance of water’s many-body characteristics.