# Beyond Energy: When Sustainability Dimensions Reshape LLM Serving Decisions

arXiv: https://arxiv.org/abs/2609.35569 · HTML: https://mlsanity.com/paper/2609.35569/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 69 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 30 of 30)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [Carbon Emissions and Large Neural Network Training](https://mlsanity.com/paper/2104.10350/) (2021)
- [RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems](https://mlsanity.com/paper/2306.03091/) (2023)
- [Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Model](https://mlsanity.com/paper/2211.02001/) (2022)
- [DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency](https://mlsanity.com/paper/2408.00741/) (2024)
- [Parallelizing the dual revised simplex method](https://mlsanity.com/paper/1503.01889/) (2015)
- [How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference](https://mlsanity.com/paper/2505.09598/) (2025)
- [Energy Considerations of Large Language Model Inference and Efficiency Optimizations](https://mlsanity.com/paper/2504.17674/) (2025)
- [Carbon-Aware Computing for Datacenters](https://mlsanity.com/paper/2106.11750/) (2021)
- [The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization](https://mlsanity.com/paper/2505.06371/) (2025)
- [Carbon Explorer: A Holistic Approach for Designing Carbon Aware Datacenters](https://mlsanity.com/paper/2201.10036/) (2022)
- [Clover: Toward Sustainable AI with Carbon-Aware Machine Learning Inference Service](https://mlsanity.com/paper/2304.09781/) (2023)
- [Intelligence per Watt: Measuring Intelligence Efficiency of Local AI](https://mlsanity.com/paper/2511.07885/) (2025)
- [Towards Sustainable Large Language Model Serving](https://mlsanity.com/paper/2501.01990/) (2024)
- [Unveiling Environmental Impacts of Large Language Model Serving: A Functional Unit View](https://mlsanity.com/paper/2502.11256/) (2025)
- [Agentic Coding in the Wild: Characterizing GitHub Copilot Traces at Production Scale](https://mlsanity.com/paper/2608.00101/) (2026)
- [Systematic Characterization of LLM Quantization: A Performance, Energy, and Quality Perspective](https://mlsanity.com/paper/2508.16712/) (2025)
- [EnsembleCI: Ensemble Learning for Carbon Intensity Forecasting](https://mlsanity.com/paper/2505.01959/) (2025)
- [A Dataset for Research on Water Sustainability](https://mlsanity.com/paper/2405.17469/) (2024)
- [Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing](https://mlsanity.com/paper/2506.22773/) (2025)
- [WaterWise: Co-optimizing Carbon- and Water-Footprint Toward Environmentally Sustainable Cloud Computing](https://mlsanity.com/paper/2501.17944/) (2025)
- [When Servers Meet Species: A Fab-to-Grave Lens on Computing's Biodiversity Impact](https://mlsanity.com/paper/2506.20442/) (2025)
- [ThirstyFLOPS: Water Footprint Modeling and Analysis Toward Sustainable HPC Systems](https://mlsanity.com/paper/2510.00471/) (2025)
- [From Cradle to Cloud: A Life Cycle Review of AI's Environmental Footprint](https://mlsanity.com/paper/2605.05416/) (2026)
- [The Sunk Carbon Fallacy: Rethinking Carbon Footprint Metrics for Effective Carbon-Aware Scheduling](https://mlsanity.com/paper/2410.15087/) (2024)
- [Cache Your Prompt When It's Green: Carbon-Aware Caching for Large Language Model Serving](https://mlsanity.com/paper/2505.23970/) (2025)
- [BIRDS: Characterizing and Understanding Biodiversity Impact of Large Language Model Serving](https://mlsanity.com/paper/2605.27480/) (2026)

## Abstract (arXiv)

Large language model (LLM) serving has environmental impacts across energy consumption, carbon emission, water consumption, and biodiversity loss. Yet these dimensions are largely evaluated in isolation, leaving it unclear when and how they lead to different optimization decisions. We present PRISM, a unified framework for characterizing and optimizing LLM serving across energy, carbon, water, and biodiversity impacts. Our analysis reveals a fundamental distinction: computing configurations determine energy consumption, whereas where and when LLM serving is deployed determine its carbon, water, and biodiversity impacts. Under a fixed deployment choice and operational-only accounting, all dimensions preserve the same energy-based configuration ranking. Deployment rankings can diverge across dimensions, while embodied impacts can break configuration invariance when they exceed a lifecycle crossover boundary. PRISM identifies these conditions, quantifies cross-dimensional regrets, and balances the four dimensions. In regional-routing experiments, PRISM reduces median worst-case regret by 50.2% relative to the strongest baseline.