A multi-location Experiment was conducted using 28 genotypes and two checks during the main cropping seasons of 2017 and 2018 at Kulumsa, Dhera, Melkasa, Asasa, Ilala, and Geregera in moisture-stress areas of Ethiopia to evaluate the genotype-Environment interaction effect and grain yield stability of bread wheat genotypes. The experiment was arranged in alpha lattice design replicated three times. The stability analysis was carried out for grain yield using R software. ANOVA showed a highly significant difference among genotypes and genotype×environmental interaction for traits; days to heading, days to maturity, plant height, grain yield, 1000 kernel weight. The average mean yield of the genotypes across the environments was 4.5 with the range of 3.5–5.3 t ha-1. The genotypes ETBW 9578, ETBW 9565, ETBW 9570 and ETBW 9571 were top-yielding, ideal, and stable across the ten environments. However ETBW 9589 and Ogolcho were the low-yielding genotypes. The “which-won-where” graph categorizes the genotypes into four groups. The locations within the same group were closely correlated and share the same winning(vertex) genotype. E7 was ideal that has both discriminating abilities of the genotypes and representative of the other test environments. ETBW 9578 gave 37.7% and 39.1% yield advantage over the standard check Kingbird and local check Ogolcho, respectively. As a result the genotype ETBW 9578 was released in 2020 and designated by the variety name “Dursa”. Thus, it was concluded that the Dursa variety could be used as a stable variety for cultivation.