The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference

Zeyu Yang, Wesley Armour
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:70498-70514, 2025.

Abstract

Deep learning has achieved significant success but poses increasing concerns about energy consumption and sustainability. Despite these concerns, there is a lack of understanding of their energy efficiency during inference. In this study, we conduct a comprehensive analysis of the inference energy consumption of 1,200 ImageNet classification models—the largest evaluation of its kind to date. Our findings reveal a steep decline in accuracy gains relative to the increase in energy usage, highlighting sustainability concerns in the pursuit of marginal improvements. We identify key factors contributing to energy consumption and demonstrate methods to improve energy efficiency. To promote more sustainable AI practices, we introduce an energy efficiency scoring system and develop an interactive web application that allows users to compare models based on accuracy and energy consumption. By providing extensive empirical data and practical tools, we aim to facilitate informed decision-making and encourage collaborative efforts in the development of energy-efficient AI technologies.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-yang25b, title = {The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference}, author = {Yang, Zeyu and Armour, Wesley}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {70498--70514}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/yang25b/yang25b.pdf}, url = {https://proceedings.mlr.press/v267/yang25b.html}, abstract = {Deep learning has achieved significant success but poses increasing concerns about energy consumption and sustainability. Despite these concerns, there is a lack of understanding of their energy efficiency during inference. In this study, we conduct a comprehensive analysis of the inference energy consumption of 1,200 ImageNet classification models—the largest evaluation of its kind to date. Our findings reveal a steep decline in accuracy gains relative to the increase in energy usage, highlighting sustainability concerns in the pursuit of marginal improvements. We identify key factors contributing to energy consumption and demonstrate methods to improve energy efficiency. To promote more sustainable AI practices, we introduce an energy efficiency scoring system and develop an interactive web application that allows users to compare models based on accuracy and energy consumption. By providing extensive empirical data and practical tools, we aim to facilitate informed decision-making and encourage collaborative efforts in the development of energy-efficient AI technologies.} }
Endnote
%0 Conference Paper %T The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference %A Zeyu Yang %A Wesley Armour %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-yang25b %I PMLR %P 70498--70514 %U https://proceedings.mlr.press/v267/yang25b.html %V 267 %X Deep learning has achieved significant success but poses increasing concerns about energy consumption and sustainability. Despite these concerns, there is a lack of understanding of their energy efficiency during inference. In this study, we conduct a comprehensive analysis of the inference energy consumption of 1,200 ImageNet classification models—the largest evaluation of its kind to date. Our findings reveal a steep decline in accuracy gains relative to the increase in energy usage, highlighting sustainability concerns in the pursuit of marginal improvements. We identify key factors contributing to energy consumption and demonstrate methods to improve energy efficiency. To promote more sustainable AI practices, we introduce an energy efficiency scoring system and develop an interactive web application that allows users to compare models based on accuracy and energy consumption. By providing extensive empirical data and practical tools, we aim to facilitate informed decision-making and encourage collaborative efforts in the development of energy-efficient AI technologies.
APA
Yang, Z. & Armour, W.. (2025). The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:70498-70514 Available from https://proceedings.mlr.press/v267/yang25b.html.

Related Material