Abstract:The rise in deployment of large language models has driven a surge in GPU demand and datacenter scaling, raising concerns about electricity use, grid stress, and the impacts of modern AI workloads. Distillation is often promoted as one of the most effective paths to obtain cheaper, more efficient models, yet these claims rarely account for the full end-to-end energy and resource costs, including crucial teacher-side workloads such as data generation, logit caching, and evaluation. We present a comprehensive energy accounting framework that measures the complete computational cost of distillation pipelines via detailed stage-wise tracking of GPU device power consumption. In our experiments, we separate and log empirical energy use across distinct phases and systematically measure the energy and emissions of two common distillation methods: the classic logit-based knowledge distillation and synthetic-data supervised fine-tuning, constructing energy-quality Pareto frontiers that expose the previously ignored costs. From these measurements and analyses, we derive practical design rules for selecting distillation methods and hyperparameters under energy and budget constraints, and release an open-source measurement harness and accounting protocol to provide a standardized foundation for comparable, reproducible distillation research, explicitly accountable for complete pipeline energy impact.
| Comments: | Accepted to the 43rd International Conference on Machine Learning (ICML 2026). 11 pages, 6 figures |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.13981 [cs.LG] |
| (or arXiv:2605.13981v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.13981 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Katherine Lambert [view email]
[v1]
Wed, 13 May 2026 18:01:02 UTC (171 KB)