Soybean [Glycine max (L.) Merrill] is major oilseed crop globally. It is grown in diverse agro-ecological conditions and the performance of quantitative traits often varies due to significant genotype × environment interaction (GEI) therefore, the integration of yield and stability is one of the common objective of soybean breeding. The present investigation was carried out to study genotype × environment interaction (GEI) through GGE biplot and AMMI analysis over four environments (Majhera, Palampur, Bajaura and Almora) with 32 genetically diverse genotypes for four traits viz., grain yield, days to 50% flowering, days to maturity and 100 seed weight under rainfed condition of North Western Himalayan hills using randomised complete block design. The analysis of variance revealed that environments (E), genotypes (G) and genotype × environment interactions (GEI) accounted about 19.61%, 26.18% and 40.71% of the total variation, respectively. GGE biplot graphically displayed interrelationships between test locations as well as genotypes and facilitated visual comparisons through two-dimensional biplot between the first two principal components (PCI and PCII) which explained 74.40% variation for grain yield, 91.98% for days to 50% flowering, 83.27% for days to maturity and 84.68% for 100 seed weight. The GGE biplot suggested suitability of all the four test locations to be used for multi-location trials on the basis of discrimination ability and representativeness. Genotypes, 'C 17' ('PS 1556') was found the best performing genotypes in terms of grain yield followed by 'C 11' ('VLS 89'), 'C 4' ('PS 1550') and 'C 10' ('DS 3102') whereas, in terms of high grain yield and stability both 'C 11' ('VLS 89') was found as the ideal genotype. In the test locations Majhera, Palampur and Almora, winning genotypes for grain yield were 'C 17' ('PS 1556') and 'C 11' ('VLS 89') while, 'C 34' ('VLS 59') was the winning genotype at Bajaura.