Summary of Large Language Model Performance Benchmarking on Mobile Platforms: a Thorough Evaluation, by Jie Xiao et al.
Large Language Model Performance Benchmarking on Mobile Platforms: A Thorough Evaluation
by Jie Xiao, Qianyi Huang, Xu Chen, Chen Tian
First submitted to arxiv on: 4 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a comprehensive measurement study on the deployment of lightweight large language models (LLMs) on commercial-off-the-shelf mobile devices. The researchers evaluate various metrics, including token throughput, latency, battery consumption, resource utilization, DVFS strategies, and inference engines, to understand the current landscape of LLM deployment on mobile platforms. They also analyze how hardware capabilities and system dynamics affect on-device LLM performance, providing insights for developers to identify and address bottlenecks. Additionally, the study provides comprehensive comparisons across different mobile SoCs from major vendors, highlighting their performance differences in handling LLM workloads. The research aims to provide a detailed understanding of the challenges and opportunities in deploying LLMs on mobile devices, which can inform both the development of on-device LLMs and the design of future mobile system architecture. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how we can use special kinds of artificial intelligence (AI) models called large language models (LLMs) on our phones. People are worried about keeping their personal data private when using these AI models, so it’s better to run them directly on the phone instead of sending data to a server somewhere else. The researchers looked at how well some lightweight LLMs can work on different types of mobile devices and what makes them fast or slow. They want to help people who make these models and the companies that design our phones understand how they can make better choices. |
Keywords
» Artificial intelligence » Inference » Token
