Performance Analysis of Lightweight LLMs on Heterogeneous Jetson Platforms

Abstract

Recently, various studies have been actively proposed to enable the deployment, efficient inference, and training of lightweight large language models (LLMs) in edge device environments. In this paper, we measure and quantitatively analyze the inference and training performance of lightweight LLMs across heterogeneous NVIDIA Jetson platforms. The results of this study can serve as a foundational reference for designing edge-intelligent frameworks that provide generative AI services in IoT environments.

Publication
Korean Institute of Communications and Information Sciences (KICS) 2025
Daeun Choi
Daeun Choi
Backend Developer

A backend developer majoring in IT Information Engineering at Jeonbuk National University.