Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2511.08575 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911259690008576 |
|---|---|
| author | Fu, Zhenxiao Fan, Chen Jiang, Lei |
| author_facet | Fu, Zhenxiao Fan, Chen Jiang, Lei |
| contents | 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 |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_08575 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CO2-Meter: A Comprehensive Carbon Footprint Estimator for LLMs on Edge Devices Fu, Zhenxiao Fan, Chen Jiang, Lei Hardware Architecture 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 |
| title | CO2-Meter: A Comprehensive Carbon Footprint Estimator for LLMs on Edge Devices |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2511.08575 |