arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.08575cs.AR
收藏

CO2-Meter:面向边缘设备大语言模型的全面碳足迹估算器

CO2-Meter: A Comprehensive Carbon Footprint Estimator for LLMs on Edge Devices

Indiana University Bloomington(印第安纳大学布卢明顿分校)

Zhenxiao Fu, Chen Fan, Lei Jiang

AI总结:

针对边缘设备LLM部署的碳挑战及现有估算器的不足,提出CO2-Meter统一框架,通过外设能耗模型、GNN预测器等实现更精准的运行与隐含碳估算,助力可持续LLM设计。

摘要:

大语言模型(LLMs)已变革自然语言处理(NLP),但在边缘设备部署面临巨大碳挑战。现有估算器不完整,忽略外设能耗、不同的预填充/解码行为及系统级芯片(SoC)设计复杂性。本文提出CO2-Meter,一个估算LLM边缘推理中运行和隐含碳的统一框架。贡献包括:(1)基于方程的外设能耗模型和数据集;(2)含阶段特异性LLM能耗数据的图神经网络(GNN)预测器;(3)用于SoC瓶颈分析的单元级隐含碳模型;(4)验证显示其精度优于现有方法。案例研究表明CO2-Meter可有效识别碳热点,指导边缘平台可持续LLM设计。源码:https://github.com/fuzhenxiao/CO2-Meter

英文摘要:

LLMs have transformed NLP, yet deploying them on edge devices poses great carbon challenges. Prior estimators remain incomplete, neglecting peripheral energy use, distinct prefill/decode behaviors, and SoC design complexity. This paper presents CO2-Meter, a unified framework for estimating operational and embodied carbon in LLM edge inference. Contributions include: (1) equation-based peripheral energy models and datasets; (2) a GNN-based predictor with phase-specific LLM energy data; (3) a unit-level embodied carbon model for SoC bottleneck analysis; and (4) validation showing superior accuracy over prior methods. Case studies show CO2-Meter's effectiveness in identifying carbon hotspots and guiding sustainable LLM design on edge platforms. Source code: https://github.com/fuzhenxiao/CO2-Meter

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑