In this paper, a new stochastic technique known as variable Particle Swarm - Grey Wolf Optimizer (vPSOGWO), a combination of metaheuristic algorithms based swarm intelligence with distinct capacities for exploration and exploitation is employed. To evaluate the efficacy of the hybrid algorithm, joint and individual inversion of various amount of synthetic datasets and finally applied on field example over various geological terrains namely for individual inversion of DC data from Digha, India and New Brunswick of Canada; MT sounding data from Sundar Pahari, Dhanbad and Puga valley, Ladakh of India and for joint inversion of DC and MT sounding data from South Central Australia. Furthermore, a posterior Bayesian probability density function using 1000 models has been computed to estimate a mean global model and uncertainty assessment. We examined the inverted results, which indicate that the results of the vPSOGWO have been shown to be more accurate than those of the PSO, GWO, and state-of-the-art variant of classic approaches. Additionally, geological significance of crustal thickness of approximately 76.581.96 km was resolved over Puga-valley, India, and is in good agreement with published data. As a result, the new approach greatly reduces uncertainty and enhances model resolution, bringing them closer to actual models.