paper-with-me

Papers

Sustainable LLM Inference for Edge AI: Evaluating Quantized LLMs for Energy Efficiency, Output Accuracy, and Inference Latency

2025-04-04 · Erik Johannes Husom, Arda Goknil, Merve Astekin, Lwin Khin Shar, Andre Kåsen, Sagar Sen, Benedikt Andreas Mithassel, Ahmet Soylu

Deploying Large Language Models (LLMs) on edge devices presents significant challenges due to computational constraints, memory limitations, inference speed, and energy consumption. Model quantization has emerged as a key technique to enable efficient LLM inference by reducing model size and computational overhead. In this study, we conduct a comprehensive analysis of 28 quantized LLMs from the Ollama library, which applies by default Post-Training Quantization (PTQ) and weight-only quantization techniques, deployed on an edge device (Raspberry Pi 4 with 4GB RAM). We evaluate energy efficiency, inference performance, and output accuracy across multiple quantization levels and task types. Models are benchmarked on five standardized datasets (CommonsenseQA, BIG-Bench Hard, TruthfulQA, GSM8K, and HumanEval), and we employ a high-resolution, hardware-based energy measurement tool to capture real-world power consumption. Our findings reveal the trade-offs between energy efficiency, inference speed, and accuracy in different quantization settings, highlighting configurations that optimize LLM deployment for resource-constrained environments. By integrating hardware-level energy profiling with LLM benchmarking, this study provides actionable insights for sustainable AI, bridging a critical gap in existing research on energy-aware LLM deployment.

📄 PDF Abstract BibTeX arXiv:2504.03360

Code (0)

등록된 구현이 없습니다.

Tasks

BenchmarkingGSM8KHumanEvalQuantizationRaspberry Pi 4TruthfulQA

Similar Papers 제목 키워드 기반

On the Compressibility of Quantized Large Language Models

2024-03-03 · Yu Mao, Weilan Wang, Hongchao Du, Nan Guan 외

Deploying Large Language Models (LLMs) on edge or mobile devices offers significant benefits, such as enhanced data privacy and real-time processing capabilities. However, it also faces critical challenges due to the sub…

Data CompressionQuantization

Quantized deep learning models on low-power edge devices for robotic systems

2019-11-30 · Anugraha Sinha, Naveen Kumar, Murukesh Mohanan, MD Muhaimin Rahman 외

In this work, we present a quantized deep neural network deployed on a low-power edge device, inferring learned motor-movements of a suspended robot in a defined space. This serves as the fundamental building block for t…

F-BFQ: Flexible Block Floating-Point Quantization Accelerator for LLMs

2025-10-15 · Jude Haris, José Cano arxiv

Large Language Models (LLMs) have become increasingly prominent for daily tasks, from improving sound-totext translation to generating additional frames for the latest video games. With the help of LLM inference framewor…

Evaluating Quantized Large Language Models for Code Generation on Low-Resource Language Benchmarks

2024-10-18 · Enkhbold Nyamsuren

Democratization of AI is an important topic within the broader topic of the digital divide. This issue is relevant to LLMs, which are becoming popular as AI co-pilots but suffer from a lack of accessibility due to high c…

Code GenerationGPUQuantization

A Comprehensive Evaluation of Quantization Strategies for Large Language Models

2024-02-26 · Renren Jin, Jiangcun Du, Wuwei Huang, Wei Liu 외

Increasing the number of parameters in large language models (LLMs) usually improves performance in downstream tasks but raises compute and memory costs, making deployment difficult in resource-limited settings. Quantiza…

Language ModelingLanguage ModellingQuantization