Please use this identifier to cite or link to this item:
http://hdl.handle.net/20.500.12188/34711| Title: | Small Prompts, Big Energy and CO2 Impact: Benchmarking Ollama LLMs on CPU and GPU | Authors: | Kolovska, Ana Gusev, Marjan Mileski, Dimitar |
Keywords: | LLM , Ollama , Energy Efficiency , Carbon Footprint , Electricity Consumption , CPU Inference , GPU Inference , Data Centers , Environmental Impact | Issue Date: | 25-Nov-2025 | Publisher: | IEEE | Conference: | 2025 33rd Telecommunications Forum (TELFOR) | Abstract: | Energy efficiency is a crucial challenge when deploying Large Language Models (LLMs). Electricity usage and related CO2 emissions can differ greatly depending on model architecture, parameter size, prompt length, and inference hardware. In this study, we evaluate 31 popular Ollama models across CPU and GPU inference, resulting in 60 testing scenarios. Energy and carbon metrics were gathered using the NVML and CodeCarbon libraries, providing insights into the environmental impact of LLM inference in data center settings. | Description: | Accepted version | URI: | http://hdl.handle.net/20.500.12188/34711 | DOI: | 10.1109/telfor67910.2025.11314365 |
| Appears in Collections: | Faculty of Computer Science and Engineering: Conference papers |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| Small Prompts, Big Energy and CO2 Impact Benchmarking Ollama LLMs on CPU and GPU - accepted version.pdf | 227.26 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.