Skip to content
Preprint

Multi-agent LLMs on the Cloud-Edge-IoT Continuum: A Systematic Mapping Study on Architectures, Deployment, and Evaluation

Oct 2026 · 0 citations · 76 references
Computer Science

TL;DR

A systematic mapping study of multi-agent LLM systems organized and deployed across the cloud-edge-IoT continuum to give deployment guidance for common IoT constraints, and identify open problems.

Abstract

LLM applications are increasingly built as systems of cooperating components rather than as single prompt-response pipelines. IoT applications add constraints of their own, because sensing, reasoning, control, and actuation are spread across devices, edge nodes, and cloud services to meet latency, privacy, and bandwidth requirements. How multi-agent LLM systems are organized and deployed across this cloud-edge-IoT continuum has not been mapped systematically. We therefore conducted a systematic mapping study. From 831 records retrieved up to March 2026, we selected 45 primary studies and classified each one by architecture family, deployment pattern, application domain, and evaluation practice. We also recorded how each system was evaluated and whether its code or data are public. Five architecture families recur in the corpus. Specialized agent teams run next to sensors, planners delegate subtasks to workers, and small local models cooperate with large cloud models. A fourth group federates model adaptation across devices, and a large fifth group builds the supporting infrastructure for offloading, communication, memory, and serving. Deployment relies on six recurring patterns, ranging from local-only execution to control loops gated by policy checks or human approval. Twenty-four of the 45 studies target IoT, industrial, smart-home, healthcare, vehicular, or robotics settings, and 29 evaluate on physical hardware, where open models below 8B parameters dominate. Based on the mapping, we list the reporting items needed to make future evaluations comparable, give deployment guidance for common IoT constraints, and identify open problems.

View source

Similar papers

#reinforcement learning Review Open access Sep 2026

A comprehensive survey of Edge-Fog-Cloud-IoT architectures: Intelligence, Security, and Sustainability

This survey provides an overview of Edge-Fog-Cloud-IoT architectural models and the technologies that have driven them during 2020–2026 and presents a holistic framework and a five-layer reference model that encompass the functional, performance, and trust aspects of distributed systems.

Patrick Effraim, Micheal Mensah, Bismark Budu · 0 citations
Conference Open access Sep 2026

Multi-Agent AI Architecture for Adaptive Service Interoperability and Autonomous Data Integration Across IoT–Cloud Platforms

The proliferation of heterogeneous Internet of Things (IoT) devices and multi-vendor cloud platforms has introduced significant challenges in achieving seamless service interoperability and autonomous data integration. Conventional centralized middleware and rule-based gateways struggle to scale with dynamic device pop...

S. Divvela · 0 citations
#edge computing Review Open access Sep 2026

Publish/Subscribe for Edge Intelligence: Systematic Review and Future Prospects

This survey provides a roadmap for advancing pub/sub mechanisms to meet the demands of interactive distributed computing environments, and proposes the integration of agentic AI to enhance adaptability and autonomy.

Alaa Saleh, Sasu Tarkoma, Susanna Pirttikangas et al. · 6 citations
Review Open access Oct 2026

Architectural Foundations for Multi-Agent Internet of Robotic Things: A Literature Review

The Internet of Robotic Things (IoRT) integrates robotics, Internet of Things (IoT), distributed computing, and intelligent systems to enable heterogeneous physical and virtual entities to perceive, communicate, collaborate, and act in dynamic environments. However, the heterogeneity and distribution of these entities...

F. Mota, Eric Rohmer · 0 citations
Open access Nov 2026

The Trade-Offs of Intelligent Orchestration in Kubernetes: A Multi-Infrastructure Case Study with the CODECO Toolkit

IoT monitoring systems, from traffic and pedestrian sensing in smart cities to building energy metering and industrial robot telemetry, increasingly run as containerised microservices on Kubernetes (K8s) clusters that span constrained edge nodes and cloud servers. Cognitive orchestration layers extend K8s with network-...

Georgios Koukis, Ioannis Dermentzis, V. Tsaoussidis et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.