Genotype × Environment interaction of 17 early duration rice genotypes tested over four seasons was analyzed to identify stable high yielding genotypes. Genotypes were grown in a randomized complete block design with three replications. Genotype × environment interaction (GEI) was analyzed following Additive Main effects and Multiplicative Interaction (AMMI) as well as regression models. AMMI analysis of variance showed highly significant genotype and environment mean squares. First two interaction principal component axes (IPCA) cumulatively explained 93.76% of total interaction effects. Integrating biplot display and genotypic stability statistics enabled four groupings of genotypes based on similarities in their performance across environments. The biplot generated using genotypes and environmental scores for first two IPCAs revealed positioning of the four genotype groups (GG) into three sectors of the biplot. Among them, three genotype in GG-3 (G-6, G-13 and G-15) exhibited high yields across environments, low IPCA-1 scores, low stability index (Di) values, unit regression coefficient and minimum deviation from regression. Hence these genotypes were recognized as possessing stable high yielding attributes. Although, both AMMI model and regression models were equally potential in partitioning GEI, AMMI analysis and biplot display was more informative in differentiating the genotype response over environments and recognizing the most discriminating environments.