A particle swarm inspired multi-elitist artificial bee colony algorithm for real-parameter optimization. The Artificial Bee Colony (ABC) algorithm is one of the most recent swarm intelligence based algorithms which simulates the foraging behavior of honey bee colonies. In this work, a particle swarm inspired multi-elitist ABC algorithm named PS-MEABC is proposed and applied for real-parameter optimization. In this modified version, the global best solution and an elitist randomly selected from the elitist archive are used to modify parameters of each food source in either onlooker bees or employed bees phases. PS-MEABC is compared with 5 state-of-the-art swarm based algorithms on CEC05 and BBOB12 benchmark functions in terms of four metrics: the mean error, the best error, the success rate ($SR$) and the expected running time ($ERT$). Wilcoxon signed ranks test results on the mean and the best error show that the performance of PS-MEABC is significantly better than or at least similar to these algorithms, and PS-MEABC has wider application range in terms of the success rate and faster convergence speed in terms of the expected running time. Our algorithm is comparable to its competitors with a fewer control parameters to be tuned.