Metraipolis › archive › dc_78ba99ddd205
How Do Prompt Variations Affect Energy Consumption in On-Device LLMs? — arXiv:2609.01798
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.01798
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T04:16:30.937Z
- SHA-256 of the archived text
d849fc8643d09b40789bb462c29440a85351c7bc45912530320191025a9934b8- Stable URL of this page
https://metraipolis.com/documento/dc_78ba99ddd205
How to cite
How Do Prompt Variations Affect Energy Consumption in On-Device LLMs? — arXiv:2609.01798 — arXiv (en), https://arxiv.org/abs/2609.01798 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_78ba99ddd205 (Metraipolis archive dc_78ba99ddd205, sha256 d849fc8643d09b40789bb462c29440a85351c7bc45912530320191025a9934b8)
BibTeX
@misc{dc78ba99ddd205,
title = {How Do Prompt Variations Affect Energy Consumption in On-Device LLMs? — arXiv:2609.01798},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.01798},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_78ba99ddd205; sha256 d849fc8643d09b40789bb462c29440a85351c7bc45912530320191025a9934b8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01798. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) are increasingly deployed on mobile devices, making energy efficiency a key deployment constraint, yet the energy impact of prompt design remains underexplored. This paper aims to understand how two prompt properties, cognitive load and phrasing pattern, shape the energy behavior of on-device LLM inference. We conduct a broad empirical study covering prompt properties, datasets, models, and devices, with phase-level profiling that separates prefill and decode energy. We find that cognitive load primarily affects the energy cost per token, while phrasing pattern affects energy largely through token usage. Our energy-quality analysis further shows that prompt design reshapes the attainable frontier differently across models, highlighting the need for model-aware prompt design in energy-efficient on-device LLM inference. Code, datasets, and scripts are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.01798 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T03:26:49.761Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).