mlsanityData to 14 Oct 2026

MARLIN: Multi-Agent Game-Theoretic Reinforcement Learning for Sustainable LLM Inference in Cloud Datacenters

arXiv 2605.13496 ↗First posted 13 May 2026cs.DC cs.LGcited by 1

On the map: LLM agents and tool use → Reinforcement learning for LLM agents

Abstract (arXiv)

Large Language Models (LLMs) have become increasingly prevalent in cloud-based platforms, propelled by the introduction of AI-based consumer and enterprise services. LLM inference requests in particular account for up to 90% of total LLM lifecycle energy use, dwarfing training energy costs. The rising volume of LLM inference requests is increasing environmental footprints, particularly carbon emissions and water consumption. To improve sustainability for LLM inference serving in cloud datacenter environments, we propose a novel multi-agent game-theoretic reinforcement learning framework called MARLIN to co-optimize time-to-first token (TTFT), carbon emissions, water usage, and energy costs associated with LLM inference. MARLIN demonstrates a reduction of at least 18% in TTFT, 33% in carbon emissions, 43% in water usage, and 11% in energy costs compared to state-of-the-art LLM inference management frameworks.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use. Builds on 14 of its 37 references that are in the map. First posted on arXiv 13 May 2026.

Builds on

14 of 37 references.

References · 14

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  2. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  3. 2017Hindsight Experience Replaycited by 1,136
  4. 2016An overview of gradient descent optimization algorithmscited by 1,033
  5. 2023Splitwise: Efficient generative LLM inference using phase splittingcited by 558
  6. 2024BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systemscited by 87
  7. 2025How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inferencecited by 76
  8. 2024Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flowcited by 55
  9. 2025Mind the Memory Gap: Unveiling GPU Bottlenecks in Large-Batch LLM Inferencecited by 33
  10. 2024PerLLM: Personalized Inference Scheduling with Edge-Cloud Collaboration for Diverse LLM Servicescited by 23
  11. 2024The Unseen AI Disruptions for Power Grids: LLM-Induced Transientscited by 10
  12. 2021Energy and Network Aware Workload Management for Geographically Distributed Data Centerscited by 6
2 more
  1. 2025Sustainable Carbon-Aware and Water-Efficient LLM Scheduling in Geo-Distributed Cloud Datacenterscited by 4
  2. 2024CASA: A Framework for SLO and Carbon-Aware Autoscaling and Scheduling in Serverless Cloud Computingcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Dissecting the Carbon-SLO Tension in Geo-Distributed LLM Servingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Game-Theoretic Deep Reinforcement Learning to Minimize Carbon Emissions and Energy Costs for AI Inference Workloads in Geo-Distributed Data Centers
  2. 2025GREEN-CODE: Learning to Optimize Energy Efficiency in LLM-based Code Generation
  3. 2025Addressing the sustainable AI trilemma: a case study on LLM agents and RAG
  4. 2026Balancing Sustainability And Performance: The Role Of Small-Scale LLMs In Agentic Artificial Intelligence Systems
  5. 2024SustainDC: Benchmarking for Sustainable Data Center Control
  6. 2026Carbon-Aware Routing for Function Calling in Edge-Cloud LLM Systems
  7. 2025Optimizing Large Language Models: Metrics, Energy Efficiency, and Case Study Insights
  8. 2023Enabling Intelligent Interactions between an Agent and an LLM: A Reinforcement Learning Approach