By Dah Reply AI
Ever since Google unveiled Gemini, comparisons with OpenAI's ChatGPT have been rampant. In our previous article, we dissected these Large Language Models (LLMs) by putting them through a unique challenge: making a pizza. Surprisingly, Gemini Pro emerged as the frontrunner, leveraging visual-rich content and offering valuable links, showcasing its prowess.
Today, we embark on a deeper comparison, delving into the realm of coding using Python. We posed two intricate questions to both LLMs to gauge their coding abilities. The first challenge involved uncovering all missing numbers within a given integer array ranging from 1 to 100. The second question pushed the boundaries by tasking the models to identify pairs of numbers within an array that sum up to a specified target value.
Apple MacBook Air M1 chip
The prompt:
“You are an expert Python programmer. How do you find all missing numbers in a given integer array of 1 to 100?”
Let's start by unraveling the mystery of missing numbers within a given integer array of 1 to 100. Here's how our contenders, ChatGPT and Gemini, tackled this challenge.
ChatGPT's Approach:
Gemini's Strategy:
Gemini with Numpy:
The first time when we feed the prompt with Gemini, it is using Numpy package to solve the problem. For fairness, we feed another prompt instructing the model to not use any external package.
We repeat executing the code using the approaches above with this array:
arr = [1, 2, 3, 5, 7, 8, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 98, 99]
In terms of performance, here are the average execution times of 10 function calling:
It's evident that Gemini's NumPy-based approach slightly impacts execution time
The prompt:
“You are an expert Python programmer. Find all pairs of numbers in an array that add up to a target value”
Moving on to the challenge of identifying pairs of numbers within an array that sum up to a specified target value:
ChatGPT's Approach:
Gemini's Strategy:
The avraged execution times of 10 function calling:
Both approaches offer clean, readable code and having similar execution time.
ChatGPT and Gemini showcased contrasting methodologies when addressing missing numbers.
ChatGPT opted for a simple iteration through the entire range, checking for each number's presence in the provided array. This straightforward approach, while intuitive and readable, might become less efficient with larger arrays due to its linear time complexity.
Gemini, on the other hand, employed a bit manipulation technique by creating a bitmap. This method offers a clever way to mark the presence of numbers and efficiently identify the missing ones in a single pass. The NumPy-based solution showcased the power of utilizing specialized libraries, potentially sacrificing a fraction of speed for more elegant and concise code.
When it came to identifying pairs that sum up to a target value, both ChatGPT and Gemini showcased a clear understanding of the task.
ChatGPT's solution involved using a dictionary to store the difference between each number and the target. This allowed for a quick lookup to identify complementary pairs. Its simplicity and ease of understanding make it an excellent choice for smaller arrays.
Meanwhile, Gemini took a slightly different approach by utilizing a dictionary to store seen numbers along with their indices. This enabled quick access to the complement needed to form a pair. Although equally effective, this method might offer a slight advantage in scenarios with larger arrays due to its implementation of indices.
While evaluating these coding challenges, it's evident that the choice of approach often depends on various factors such as array size, computational efficiency, and code readability.
In the arena of coding challenges, both ChatGPT and Gemini demonstrate impressive capabilities. ChatGPT shines with its clear and understandable code, providing an excellent foundation for smaller tasks and scenarios where readability is paramount. Conversely, Gemini's strategies exhibit prowess in optimizing for efficiency and scalability, showcasing the power of leveraging advanced techniques and libraries.
In essence, the choice between these approaches often boils down to the specific needs of the task at hand—whether it's prioritizing readability, optimizing for performance, or harnessing specialized tools.
Stay tuned for further explorations and insights into the coding prowess of these remarkable language models!
Discover Dah Reply, a leading Malaysia Chatbot Startup dedicated to crafting high quality chatbot solutions tailored for business growth and high ROI. Experience the difference by kickstarting your business enhancement today with a free demo.
Enter your email to receive our latest newsletter.
Compare pizza-making guidance from Google LLM Gemini Pro and OpenAI's ChatGPT in our detailed analysis. Explore Gemini Pro's multimedia-rich instructions against ChatGPT's text-based approach. Find the perfect recipe for crafting homemade pizza with unique insights into each method!
Dive into Google's groundbreaking Gemini AI, a game-changer in technology evolution. Explore its three versions, from Ultra to Nano, showcasing unparalleled performance and transformative abilities in diverse domains. Discover Gemini's potential from revolutionizing scientific data extraction to redefining coding landscapes. Join us as we uncover the future of AI with Google's cutting-edge innovation.