How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912725693628416 |
|---|---|
| author | Jegham, Nidhal Abdelatti, Marwan Koh, Chan Young Elmoubarki, Lassad Hendawi, Abdeltawab |
| author_facet | Jegham, Nidhal Abdelatti, Marwan Koh, Chan Young Elmoubarki, Lassad Hendawi, Abdeltawab |
| contents | This paper introduces an infrastructure-aware benchmarking framework for quantifying the environmental footprint of LLM inference across 30 state-of-the-art models in commercial datacenters. The framework combines public API performance data with company-specific environmental multipliers and statistical inference of hardware configurations. We additionally utilize cross-efficiency Data Envelopment Analysis (DEA) to rank models by performance relative to environmental cost and provide a dynamically updated dashboard that visualizes model-level energy, water, and carbon metrics. Results show the most energy-intensive models exceed 29 Wh per long prompt, over 65 times the most efficient systems. Even a 0.42 Wh short query, when scaled to 700M queries/day, aggregates to annual electricity comparable to 35{,}000 U.S. homes, evaporative freshwater equal to the annual drinking needs of 1.2M people, and carbon emissions requiring a Chicago-sized forest to offset. These findings highlight a growing paradox: as AI becomes cheaper and faster, global adoption drives disproportionate resource consumption. Our methodology offers a standardized, empirically grounded basis for sustainability benchmarking and accountability in AI deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_09598 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference Jegham, Nidhal Abdelatti, Marwan Koh, Chan Young Elmoubarki, Lassad Hendawi, Abdeltawab Computers and Society Artificial Intelligence This paper introduces an infrastructure-aware benchmarking framework for quantifying the environmental footprint of LLM inference across 30 state-of-the-art models in commercial datacenters. The framework combines public API performance data with company-specific environmental multipliers and statistical inference of hardware configurations. We additionally utilize cross-efficiency Data Envelopment Analysis (DEA) to rank models by performance relative to environmental cost and provide a dynamically updated dashboard that visualizes model-level energy, water, and carbon metrics. Results show the most energy-intensive models exceed 29 Wh per long prompt, over 65 times the most efficient systems. Even a 0.42 Wh short query, when scaled to 700M queries/day, aggregates to annual electricity comparable to 35{,}000 U.S. homes, evaporative freshwater equal to the annual drinking needs of 1.2M people, and carbon emissions requiring a Chicago-sized forest to offset. These findings highlight a growing paradox: as AI becomes cheaper and faster, global adoption drives disproportionate resource consumption. Our methodology offers a standardized, empirically grounded basis for sustainability benchmarking and accountability in AI deployment. |
| title | How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference |
| topic | Computers and Society Artificial Intelligence |
| url | https://arxiv.org/abs/2505.09598 |