The effective over-allocation of Virtual Machines (VMs) within cloud data centres has a profound effect on the demand on power, time deferment in addition to resource usage. In the current study, we suggest a Hybrid Multi-Objective Swarm Intelligence (HMOSI) model combining the Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) models to improve VM placement strategies. Our model optimizes three incompatibility objectives: it tries to reduce power, to reduce network latency and maximizes resource utilization. We compare the HMOSI model to well-known methods such as First Fit Decreasing (FFD), PSO-only and ACO-only model on real-world data traces as well as Cloud Sim simulation trace across different workloads, with the results showing the advantages of the proposed solution. The HMOSI-based results indicate that the trade-offs between these objectives can be successfully balanced and results in a significant improvement in power consumption (18.9%), a significant reduction in the network delay (35.6%), and resource utilization (11.5%) over existing solutions. This signifies that the HMOSI can enhance the level of operation efficiency in large-scale dynamic cloud environments where workload variable and flexibility live and breathe. Our results indicate that the hybrid character of the model, where the global exploration power of the PSO is in collaboration with the local exploitation potential of the ACO, provides a durable and scalable approach to VM placement optimization in cloud data centres.