With the deep integration of 5G-Advanced (5G-A) and Artificial Intelligence (AI) technologies, the Multi-Agent Based 5G-A Network is becoming a crucial mobile infrastructure for enabling the deployment of intelligent applications such as embodied robots, AI intelligent assistants, autonomous driving and so on. This project aims to build an intelligent network with a "sensing-analysis-decision-execution" closed loop through the collaborative decision-making and resource optimization of AI multi-agents, significantly enhancing the reliability, real-time performance, and scalability of intelligent applications. This solution is based on the 5G-A network and introduces a multi-agent architecture, innovatively adopting a real-time mode that combines hierarchical collaboration and peer-to-peer collaboration. It supports the dynamic expansion and modular updates of intelligent agents for operations and maintenance, optimization, energy saving, and services. By defining Agent-to-Agent interfaces for operators and equipment vendors, cross-platform compatibility is achieved, forming an open "network + computing power + AI" platform. Operators can flexibly orchestrate and provide customized network capability services based on the specific business characteristics of different scenarios, meeting the deterministic requirements of AI intelligent applications for high reliability, large uplink capacity, low inference latency, and high accuracy, enabling the mobile network to transform from a traditional "pipe provider" to an "intelligent service platform." This project provides strong mobile network capability support for AI intelligent applications, effectively promoting their commercialization. For example, embodied robots can leverage the 5G-A+AI multi-agent network capabilities to achieve dynamic motion control, visual recognition, and task planning, enabling them to autonomously adjust strategies and reduce manual intervention. The intelligent robot market is projected to reach 1.2 trillion RMB in 2025, with a compound annual growth rate of 24%, highlights this solution's potential. This solution pioneers the introduction of multi-agent technology into mobile networks, forming an AI product solution with industry characteristics that better meets the future deployment of AI intelligent services.