Optimizing AI Language Models: A Study of ChatGPT-4 vs. ChatGPT-4o

Authors

Md Nurul Absar Siddiky, University of North Carolina at Charlotte
Muhammad Enayetur Rahman, Old Dominion UniversityFollow
MD Fayaz Bin Hossen, Old Dominion UniversityFollow
Muhammad Rezaur Rahman, Agile Crafts
Md. Shahadat Jaman, Bangladesh University of Business and Technology

Document Type

Article

Publication Date

2025

DOI

10.20944/preprints202502.0066.v1

Publication Title

Preprints.org

Pages

22 pp.

Abstract

This paper presents a comparative analysis of OpenAI's GPT-4 and its optimized variant, GPT-4o, focusing on their architectural differences, performance, and real-world applications. GPT-4, built upon the Transformer architecture, has set new standards in natural language processing (NLP) with its capacity to generate coherent and contextually relevant text across a wide range of tasks. However, its computational demands, requiring substantial hardware resources, make it less accessible for smaller organizations and real-time applications. In contrast, GPT-4o addresses these challenges by incorporating optimizations such as model compression, parameter pruning, and memory-efficient computation, allowing it to deliver similar performance with significantly lower computational requirements. This paper examines the trade-offs between raw performance and computational efficiency, evaluating both models on standard NLP benchmarks and across diverse sectors such as healthcare, education, and customer service. Our analysis aims to provide insights into the practical deployment of these models, particularly in resource-constrained environments.

Comments

This is a preprint article. It has not been peer-reviewed.

Rights

© 2025 The Authors.

This open access article is published under a Creative Commons Attribution 4.0 International (CC BY 4.0) License, which permit the free download, distribution, and reuse, provided that the authors and preprint are cited in any reuse.

Original Publication Citation

Siddiky, M. N. A., Rahman, M. E., Hossen, M. F. B., Rahman, M. R., & Jaman, M. S. (2025). Optimizing AI language models: A study of ChatGPT-4 vs. ChatGPT-4o. Preprints.org. https://doi.org/10.20944/preprints202502.0066.v1

Repository Citation

Siddiky, Md Nurul Absar; Rahman, Muhammad Enayetur; Hossen, MD Fayaz Bin; Rahman, Muhammad Rezaur; and Jaman, Md. Shahadat, "Optimizing AI Language Models: A Study of ChatGPT-4 vs. ChatGPT-4o" (2025). Electrical & Computer Engineering Faculty Publications. 512.
https://digitalcommons.odu.edu/ece_fac_pubs/512

ORCID

0009-0004-0421-5540 (Rahman), 0009-0008-5109-2466 (Bin Hossen)