Four parental genotypes of okra were crossed in complete diallel design to study the direction and extent of relative heterosis and heterobeltiosis for yield and its associated traits for utilization of existing genetic diversity to develop heterotic F1 hybrids in okra. The additive genetic component (D) was significant in all studied traits except average pod weight. Nonadditive (H1 and H2) components were found to be significant in all studied traits. However, the values of the dominant effect (H1) were smaller than the D components for no. of nodes/plant, no. of pods/plant, weight of medium pods, weight of large pods, and total fresh pod yield. The maximum significant MP heterosis in the desirable direction (149.9%) was recorded for the weight of large pods/plot. The maximum significant heterobeltiosis in the desirable direction (120.1%) was recorded for the weight of small pods/plot followed by total fresh pod yield (107.4%), the weight of large pods/plot (104.9%), weight of medium pods/plot (92.1%), average pod weight (51.8%), number of pods/plant (38.4%), and plant height (34.3%). It could be concluded that plant height, average pod weight, and the number of branches could be considered for the development of elite hybrids (heterosis breeding) or inbred lines (pure line selection) in succeeding generations. Therefore, these parameters can be considered for selecting genotypes to improve the pod yield of okra. The superior crosses identified through heterosis analysis were Egyptian Balady × Line 4.1.18 (30.8 ton/ha), Line 4.1.18 × Egyptian Balady (29.8 ton/ha), Dwarf Green Long Pod × Line 4.1.18 (28.3 ton/ha), and Egyptian Balady × Dwarf Green Long Pod (27.6 ton/ha) as these crosses had high performance as well as significant and higher estimates of heterobeltiosis for fruit yield per plant and yield attributing other characters.