A Performance Study of LLM-Generated Code on Leetcode
This provides empirical evidence for developers and researchers about LLM code generation capabilities, though it is incremental in benchmarking existing models.
This study evaluated the performance of code generated by 18 Large Language Models (LLMs) on Leetcode problems, finding that LLMs produce code with comparable efficiency across models and, on average, more efficient than human-written solutions.
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.