Fifteen improved and two local genotypes of Vigna unguiculata were evaluated for their performance in grain yield, yield components and yield stability from 2016 to 2018. Experiments were laid out in a Randomised Complete Block Design, with three replications. Data were collected on grain yield and yield components and subjected to analysis of variance (ANOVA). Stability analysis for grain yield was conducted using Francis and Kannenberg's genotype-grouping technique. Results of combined ANOVA revealed highly significant (p<0.001) differences among genotypes, and across years for studied traits, except for days to physiological maturity, with non-significant yearly variations. Genotype by year interactions were non-significant except for number of seeds per pod, days to 50% flowering and hundred seed weight. Genotype, IT08K-126-19, gave the highest mean grain yield (3611 kgha -1 ), while "Akidi elu" gave the least (1695 kgha -1 ). Genotypes, IT08K-126-19, IT07K-210-1-1 and IT09K-456 gave higher and more stable grain yields. Meanwhile, IT08K-180-11, IT10K-837-1 and "Akidi elu", gave lower and unstable yields. The results revealed that sufficient variability exists among genotypes, which can be exploited in breeding programmes. Genotypes with high and stable yields can be released to farmers to boost productivity.