Abstract: A field experiment was conducted at Woreta, Metema and Pawe of North West Ethiopia during the 2008/9 and 2009/10 cropping seasons. Sixteen upland rice genotypes including the standard check were evaluated with the objective of selecting stable and high-yielding, disease-resistant and early maturing varieties. The experiment was laid out in a randomized complete block design of three replications. Combined analysis of variance revealed significant variations in genotypes for most parameters except fertile tillers per plant. Mean squares due to the genotype × location × year interaction was significant for days to aturity, 1000-seed weight and grain yield (P<0.01). The highest grain yield of 3.5t ha-1 was recorded by G15, followed by G7 (3.34t ha-1). In AMMI analysis, variations due to environments accounted for 52.48% of the treatment sum of square (SS) while genotypes and genotype × environment interaction explained 12.06 % and 35.45% of SS, respectively. In GGE biplot, G15 and G7 respectively were the best genotypes in E1, E2, and E6. The Genotype G15 was relatively stable in grain yield across environments and it also showed relatively better resistance to major rice diseases. Hence, G15 could be recommended for cultivation by the farmers and this variety should be popularized in larger scale to make use of its merits.