mlsanityData to 7 Oct 2026

Greenpixie's AI Token Methodology: Assessing the Energy, Water and CO2-eq Impact of AI Tokens for Open and Closed Weight Models

arXiv 2609.33965 ↗First posted 27 Sep 2026cs.CY cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We describe a methodology for estimating the per-token energy cost of cloud-hosted large language model (LLM) inference, separating between input (prefill) and output (decode) tokens. Graphics processing unit (GPU) energy usage is measured during inference benchmarking with open-weights models on a wide range of text-based tasks. The remaining server energy contribution from non-GPU hardware is estimated from the inference wall time. Bayesian linear regression is used to model the relationship between energy per token and LLM size, request traffic, and hardware deployment configuration. Proprietary frontier LLMs of unknown size and deployment are binned into size buckets based on naming conventions and performance priors, and the space of possible LLM configurations is sampled with Monte-Carlo methods to give a representative average energy per token and uncertainty. We also describe how these energy measurements can be used to estimate the carbon-dioxide equivalent (mathrmCO2-eq) emissions, both usage and embodied, and water consumed per token of AI inference. This methodology provides actionable data that enables reductions in cost, electricity usage, mathrmCO2-eq emitted and water consumed in cloud and Software as a Service (SaaS).

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 67 references that are in the map. First posted on arXiv 27 Sep 2026.

Builds on

32 of 67 references.

References · 32

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  4. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  5. 2016Pointer Sentinel Mixture Modelscited by 2,992
  6. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  7. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  8. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  9. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,538
  10. 2018Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarizationcited by 1,308
  11. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,143
  12. 2015Teaching Machines to Read and Comprehendcited by 1,039
20 more
  1. 2022Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Modelcited by 264
  2. 2020Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learningcited by 246
  3. 2021Sustainable AI: Environmental Implications, Challenges and Opportunitiescited by 242
  4. 2022FP8 Formats for Deep Learningcited by 187
  5. 2023Power Hungry Processing: Watts Driving the Cost of AI Deployment?cited by 186
  6. 2022The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrinkcited by 183
  7. 2023From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inferencecited by 168
  8. 2025TurboQuant: Online Vector Quantization with Near-optimal Distortion Ratecited by 116
  9. 2025How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inferencecited by 73
  10. 2024Towards Greener LLMs: Bringing Energy-Efficiency to the Forefront of LLM Inferencecited by 59
  11. 2025Energy Considerations of Large Language Model Inference and Efficiency Optimizationscited by 54
  12. 2019Energy Usage Reports: Environmental awareness as part of algorithmic accountabilitycited by 43
  13. 2025The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimizationcited by 31
  14. 2025Measuring the environmental impact of delivering AI at Google Scalecited by 31
  15. 2024Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systemscited by 27
  16. 2025TokenPowerBench: Benchmarking the Power Consumption of LLM Inferencecited by 21
  17. 2025Energy Use of AI Inference, Efficiency Pathways, and Test-Time Scalingcited by 12
  18. 2026Where Do the Joules Go? Diagnosing Inference Energy Consumptioncited by 10
  19. 2026From Tokens to Watt-hours: Analytical Energy Estimation for LLM Inference on Modern GPUscited by 3
  20. 2026Characterization of Request and Token Energy Costs for LLM Inference Workloads on GPU Platformscited by 2

Built on by

Not cited yet.