Economic Load Dispatch (ELD) is an important optimization problem in power-system operation because it seeks to allocate generation among available generating units while satisfying system demand and operating constraints. This study investigates the convergence and transmission-loss performance of a Hybrid Artificial Bee Colony–Firefly Algorithm (ABC–FA) for the ELD problem of the Nigerian 330 kV power system. Three population-based optimization approaches, namely Artificial Bee Colony (ABC), Firefly Algorithm (FA), and Hybrid ABC–FA, were implemented in MATLAB and evaluated under load-demand conditions ranging from 2000 MW to 4000 MW. The hybrid approach combines the exploration capability of ABC with the exploitation characteristics of FA. Convergence characteristics were evaluated from the variation of the objective-function value with iteration number, while transmission losses were assessed through power-balance verification. The reported convergence curves show rapid improvement during the initial search stage followed by gradual convergence as the algorithms approach their final solutions. The Hybrid ABC–FA generally demonstrated faster convergence and improved final solution characteristics compared with the individual algorithms. For transmission losses, the verified results show that the ABC and Hybrid ABC–FA methods produced physically plausible loss estimates across the investigated load levels, while two standalone FA cases at 2000 MW and 2500 MW were identified in the report as problematic and were therefore excluded from the comparative loss analysis. The study demonstrates the potential of hybrid swarm-intelligence optimization for improving the computational performance of economic dispatch in the Nigerian 330 kV power system.