Linear regression model of Eberhart and Russell is used to identify high yielding stable little millet genotypes suitable across environments. Pooled analysis of variance revealed significant genetic variability among the little millet genotypes for yield and yield attributing traits. Significant variability among environments confirms the heterogeneity in the locations for the traits. Significant genotype x environment interaction for all the traits indicated differential response of the genotypes for the traits in different locations. Among genotypes BL-6, LMNDL-4, LMNDL-3, OLM 203, VS 13, VS 15, VS 19, VS 25 and VS 6, with a regression coefficient near to unity and non-significant deviation from regression, were considered to be highly stable and suitable to all environments for fodder yield. For grain yield VS10 and VS 19 genotypes recorded regression coefficient near to unity with non-significant deviation from regression were considered to be stable in all the environments. GGE biplot model were used to evaluate stability for important traits; fodder yield and grain yield and test location representativeness in little millet genotypes. The testing locations were partitioned into two mega environments (ME) for fodder yield. ME1 was represented by Nandyal and Vizianagaram with OLM 203 as the winning genotype, while in ME 2 which was represented by Perumallapalle with LMNDL 4 performed well. Genotype VS 10 was stable for fodder yield in all the environments.For grain yield over pooled locations, BL 6 performed well at Perumallapalle location while LMNDL 5 performed well at Nandyal. Genotype VS 6 was the best genotype for Vizianagaram. The genotype VS 19 was near to origin and it was considered as stable for all the environments.