The Agentic6G Agentic-AI system based on Multi-Agent Systems (MAS) will enhance and evolve the interplay between 6G network and computing resources across the Device–Edge–Cloud continuum, potentially leveraging new 6G architectural advancements, in addition to the extreme communication capabilities beyond 5G.

For service composition, Agentic6G will allow the MAS to autonomously compose, coordinate, and optimise B5G/6G and/or vertical business microservices for autonomous service composition, deployment and management, by combining collective agentic AI’s knowledge with the support of Large Language Models (LLMs), shared memories and tools.

Agentic6G will enable efficient end-to-end Agentic Operations (AgenticOps) to support the lifecycle management of MAS. In addition, it will provide self-organising capabilities, including spontaneous agent generation, as well as self-protection, self-healing, and auto-scaling of the system.

The main outcomes will be a set of Agentic6G agents as autonomy enablers and an integrated MAS, tested, validated, and demonstrated through representative technical use cases, including a complex swarm robotics application.