Networking-Aware Energy Efficiency in Agentic AI Inference: A Survey

arXiv 2604.07857 ↗First posted 9 Apr 2026eess.SY cs.AI cs.SYcited by 0

On the map: LLM agents and tool use → Agentic AI for wireless networks

Abstract (arXiv)

The rapid emergence of Large Language Models (LLMs) has catalyzed Agentic artificial intelligence (AI), autonomous systems integrating perception, reasoning, and action into closed-loop pipelines for continuous adaptation. While unlocking transformative applications in mobile edge computing, autonomous systems, and next-generation wireless networks, this paradigm creates fundamental energy challenges through iterative inference and persistent data exchange. Unlike traditional AI where bottlenecks are computational Floating Point Operations (FLOPs), Agentic AI faces compounding computational and communication energy costs. In this survey, we propose an energy accounting framework identifying computational and communication costs across the Perception-Reasoning-Action cycle. We establish a unified taxonomy spanning model simplification, computation control, input and attention optimization, and hardware-aware inference. We explore cross-layer co-design strategies jointly optimizing model parameters, wireless transmissions, and edge resources. Finally, we identify open challenges of federated green learning, carbon-aware agency, 6th generation mobile communication (6G)-native Agentic AI, and self-sustaining systems, providing a roadmap for scalable autonomous intelligence.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Agentic AI for wireless networks, one of 16 directions in LLM agents and tool use. Builds on 55 of its 125 references that are in the map. First posted on arXiv 9 Apr 2026.

Builds on

55 of 125 references.

References · 55

2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
2023SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shotcited by 1,090
2024KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cachecited by 604
2024Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inferencecited by 487
2017Efficient Processing of Deep Neural Networks: A Tutorial and Surveycited by 427
2022DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scalecited by 400
2023Atom: Low-bit Quantization for Efficient and Accurate LLM Servingcited by 255

38 more

2020SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruningcited by 228
2024MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Casescited by 216
2023CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Servingcited by 199
2023LLM in a flash: Efficient Large Language Model Inference with Limited Memorycited by 170
2023SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inferencecited by 89
2024Loki: Low-rank Keys for Efficient Sparse Attentioncited by 87
2023EdgeMoE: Empowering Sparse Large Language Models on Mobile Devicescited by 65
2024A Survey on Inference Optimization Techniques for Mixture of Experts Modelscited by 46
2024AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruningcited by 43
2016Wireless Communications with Unmanned Aerial Vehicles: Opportunities and Challengescited by 40
2019Wireless Federated Distillation for Distributed Edge Learning with Heterogeneous Datacited by 39
2023Action-Quantized Offline Reinforcement Learning for Robotic Skill Learningcited by 34
2025Sustainable LLM Inference for Edge AI: Evaluating Quantized LLMs for Energy Efficiency, Output Accuracy, and Inference Latencycited by 32
2017Joint Trajectory and Communication Design for Multi-UAV Enabled Wireless Networkscited by 31
2025TAPAS: Thermal- and Power-Aware Scheduling for LLM Inference in Cloud Platformscited by 28
2023SwapMoE: Serving Off-the-shelf MoE-based Large Language Models with Tunable Memory Budgetcited by 27
2024CE-CoLLM: Efficient and Adaptive Large Language Models Through Cloud-Edge Collaborationcited by 25
2024SLO-aware GPU Frequency Scaling for Energy Efficient LLM Inference Servingcited by 18
2024Hybrid Heterogeneous Clusters Can Lower the Energy Consumption of LLM Inference Workloadscited by 17
2024A Queueing Theoretic Perspective on Low-Latency LLM Inference with Variable Token Lengthcited by 16
2025DSSD: Efficient Edge-Device LLM Deployment and Collaborative Inference via Distributed Split Speculative Decodingcited by 15
2024MeanCache: User-Centric Semantic Caching for LLM Web Servicescited by 14
2023ConsistentEE: A Consistent and Hardness-Guided Early Exiting Method for Accelerating Language Models Inferencecited by 13
2022Multi-Agent Collaborative Inference via DNN Decoupling: Intermediate Feature Compression and Edge Learningcited by 12
2024KVPR: Efficient LLM Inference with I/O-Aware KV Cache Partial Recomputationcited by 12
2025Token Communication in the Era of Large Models: An Information Bottleneck-Based Approachcited by 10
2025TinyServe: Query-Aware Cache Selection for Efficient LLM Servingcited by 9
2024Edge Intelligence Optimization for Large Language Model Inference with Batching and Quantizationcited by 7
2025Quality-of-Service Aware LLM Routing for Edge Computing with Multiple Expertscited by 7
2023Early-Exit Neural Networks with Nested Prediction Setscited by 6
2024Retrieval-augmented Generation for GenAI-enabled Semantic Communicationscited by 5
2019Modeling and Analysis of Energy Harvesting and Smart Grid-Powered Wireless Communication Networks: A Contemporary Surveycited by 4
2025Dynamic Quality-Latency Aware Routing for LLM Inference in Wireless Edge-Device Networkscited by 4
2023GreenCourier: Carbon-Aware Scheduling for Serverless Functionscited by 3
2025Task-Oriented Multimodal Token Transmission in Resource-Constrained Multiuser Networkscited by 3
2025Toward Sustainability-Aware LLM Inference on Edge Clusterscited by 3
2025Input Conditioned Layer Dropping in Speech Foundation Modelscited by 2
2025AGFT: An Adaptive GPU Frequency Tuner for Real-Time LLM Inference Optimizationcited by 2

50 most cited of 55 All 55 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way