Journal article · 2026 · Open access
Automated Code Generation Using Large Language Models: A Comparative Study
Research consortium archive. · Published 19 April 2026
In brief
“Automated Code Generation Using Large Language Models: A Comparative Study” is a 2026 journal article by Rimsha Naseem and Muhammad Zaki, published in Research consortium archive.. OpenAlex counts 0 citations as of 19 November 2026. It is open access: the full text is free to read at Research consortium archive..
- Citations
- 0
- Authors
- 2
Abstract
Background: The emergence of large language models (LLMs) has revolutionized automated code generation, which has the potential to increase the productivity of software development. Recent models Codex, AlphaCode, CodeGen and Code Llama have achieved mixed results on coding, but a comprehensive comparison of their effectiveness is not yet available. We attempt to fill that gap by comparing these models under several important aspects of code generation. Objective: This study aims to perform a comprehensive comparative analysis on Codex, AlphaCode, CodeGen, and Code Llama, in terms of their effectiveness in creating valid and efficient code for a variety of tasks involving algorithmic problem-solving, API invocations and web development. Performance The study is also intended to compare the models with several performance metrics, including accuracy, execution success, code efficiency, and safety. Method: The chosen LLMs were experimented on the benchmark datasets (HumanEval, MBPP). Quantitative measures in terms of accuracy, percentage of successfully executed attacks, computational efficiency, and the analysis of discriminative ability were used. Subjective measures (readability, creativity and error types) were also evaluated. A setup of prompt design and model parameters (temperature, etc.) was adjusted for a fair comparison. Results: Across all measures, Codex is found to be the best with the exception of for creative problem solving with AlphaCode performing the best. CodeGen showed a higher efficiency and a lower susceptibility rate. Code Llama came in second place, with lesser performance, but potential for improvement. Conclusion: Overall, Codex tops, but there are specific benefits offered by AlphaCode, CodeGen, and Code Llama for various use-case scenarios. In future research, better security, further optimizing of models for specialized tasks, and removing disadvantages should be the three main parts that researchers can try.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Naseem, R., & Zaki, M. (2026). Automated Code Generation Using Large Language Models: A Comparative Study. Research consortium archive.. https://rcresearcharchive.com/index.php/index/article/view/733
MLA 9
Naseem, Rimsha, and Muhammad Zaki. “Automated Code Generation Using Large Language Models: A Comparative Study.” Research consortium archive., 2026. https://rcresearcharchive.com/index.php/index/article/view/733.
Chicago
Naseem, Rimsha and Muhammad Zaki. 2026. “Automated Code Generation Using Large Language Models: A Comparative Study.” Research consortium archive.. https://rcresearcharchive.com/index.php/index/article/view/733.
Harvard
Naseem, R. and Zaki, M. (2026) ‘Automated Code Generation Using Large Language Models: A Comparative Study’, Research consortium archive.. Available at: https://rcresearcharchive.com/index.php/index/article/view/733.
IEEE
R. Naseem and M. Zaki, “Automated Code Generation Using Large Language Models: A Comparative Study,” Research consortium archive., 2026.
BibTeX
@article{naseem2026automated,
title = {{Automated Code Generation Using Large Language Models: A Comparative Study}},
author = {Naseem, Rimsha and Zaki, Muhammad},
journal = {Research consortium archive.},
year = {2026},
url = {https://rcresearcharchive.com/index.php/index/article/view/733}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- automatic code generation
- large language models
- codices
- CodeGen
- CodeLlama
- model comparison
- HumanEval
- MBPP
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Naseem, R., & Zaki, M. (2026). Automated Code Generation Using Large Language Models: A Comparative Study. Research consortium archive.. https://rcresearcharchive.com/index.php/index/article/view/733
How many times has this paper been cited?
0 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at Research consortium archive.: sign in and open the full text from this page.
Who wrote this paper?
Rimsha Naseem, Muhammad Zaki.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.