A Performance Study of LLM-Generated Code on Leetcode

Abstract

This study evaluates the efficiency of code generation by Large Language Models (LLMs) and measures their performance against human-crafted solutions using a dataset from Leetcode. We compare 18 LLMs, considering factors such as model temperature and success rate, and their impact on code performance. This research introduces a novel method for measuring and comparing the speed of LLM-generated code, revealing that LLMs produce code with comparable performance, irrespective of the adopted LLM. We also find that LLMs are capable of generating code that is, on average, more efficient than the code written by humans. The paper further discusses the use of Leetcode as a benchmarking dataset, the limitations imposed by potential data contamination, and the platform's measurement reliability. We believe that our findings contribute to a better understanding of LLM capabilities in code generation and set the stage for future optimizations in the field.

References in corpus (5)

• Code Llama: Open Foundation Models for Code
• Do Users Write More Insecure Code with AI Assistants?
• StarCoder: may the source be with you!
• ReCode: Robustness Evaluation of Code Generation Models
• TrojanPuzzle: Covertly Poisoning Code-Suggestion Models

Cited by in corpus (5)

• A Comprehensive Survey on Integrating Large Language Models with Knowledge-Based Methods
• One Size Does Not Fit All: Investigating Efficacy of Perplexity in Detecting LLM-Generated Code
• LLM-ProS: Analyzing Large Language Models' Performance in Competitive Problem Solving
• Boardwalk: Towards a Framework for Creating Board Games with LLMs
• TaskEval: Assessing Difficulty of Code Generation Tasks for Large Language Models