Twenty nine okra accessions sourced from different agro-ecological regions in Nigeria were tested for stability and performance in four environments between 2006 and 2008 using Additive main effect and multiplicative interaction (AMMI) and Genotype main effect and Genotype by Environment (GGE) models. The experiment was laid out in a Randomized Complete Block Design (RCBD) with five replications. Both AMMI and GGE biplots identified three common genotypes that were overall best in performance in relation to yield and stability. This suggests that for reliability and optimum result it is better to combine the result of the two analytical tools for yield and stability in the recommendation of genotypes to farmers. Both AMMI and GGE models identified LD88/1-8-5-2(G7), 47-4(G17) and NH88/1-8-16-2(G25) as the best accessions for cultivation across seasons because they combined stability and above average yield. Similarly both AMMI and GGE models identified Abeokuta dry season as more stable and ideal for okra genetic evaluation, while Abeokuta rainy season was the most unstable but gave better mean performance. CCN2005/1 (10.92 g/plant) and Clemson spineless (7.72 g/plant) had the least yield and less stable, hence unsuitable for cultivation across seasons. OSADEP Purple tall and NH99/9 were most adapted to Abeokuta rainy season
