Abstract. This paper presents a fast and accurate tsunami real time operational model to compute across-ocean wide-simulations completely on GPU. The spherical shallow water equations are solved using the method of characteristics and upwind cubic-interpolation, to provide high accuracy and stability. A customized, user interactive, tree based mesh 10 refinement method is implemented based on distance from the coast and focal areas to generate a memory efficient domain with resolutions of up to 50m. Three GPU kernels, specialized and optimized (wet, wall and inundation) are developed to compute the domain block mesh. Multi-GPU is used to further speed up the computation and a weighted Hilbert space filling curve is used to produce balanced work load. Hindcasting of the 2004 Indonesia tsunami is presented to validate and compare the agreement of the arrival times and main peaks at several gauges. Inundation maps are also produced for Kamala 15 and Hambantota to validate the accuracy of our model. Test runs on three Tesla P100 cards on Tsubame 3.0 could fully simulate 10 hours in just under 10 minutes wall clock time.