Claude Sonnet 3.5 vs ChatGPT-4o (2)

AI Comparison Series II

In partnership with

Welcome AI-Powered Leaders! 👋

Good morning. In today's newsletter, we will compare Claude Sonnet 3.5 vs. ChatGPT-4o in terms of business use cases. As the second version of this series, we will be focusing on visual-related outputs.

Reading Time: Approximately 6 Minutes

INTRODUCTION

AI Models Comparison

In our previous comparison newsletter, we tested the text-based outputs generated by ChatGPT-4 and Claude Sonnet 3.5 to determine which is more effective. This newsletter will focus on visual-based outputs, such as vision capabilities and graph creation. We will do this by providing identical input to both models and then analyzing the results.

TODAY’S SPONSOR

Use AI as Your Personal Assistant

Ready to embrace a new era of task delegation?

HubSpot’s highly anticipated AI Task Delegation Playbook is your key to supercharging your productivity and saving precious time.

Learn how to integrate AI technology into your processes, allowing you to optimize resource allocation and maximize output with precision and ease.

AI COMPARISON

Practical Applications to AI models

In this part, we will be testing out two use cases, which are vision capabilities and analyzing a graph for two different models. You will see the same prompts and inputs for the two models and a comparison of the outputs for both.

1.Vision Capabilities

Here is the prompt and input we will use for vision capability purposes:

Please analyze the details of this image and provide an explanation.

Here is the image we use for this test:

Additionally, in the below, first we uploaded an image and did not prompt anything to see what both AI models would do just by uploading the image, and then we prompt in order to analyze the picture. Here are the results for two models.

ChatGPT-4o Outputs:

Claude Sonnet 3.5 Outputs:

Comparison:

When we first showed the picture to both AI models without giving them any instructions, they began to examine it and give a thorough description. To sum up, Claude's review offers a more in-depth subtle account that pays attention to contrast, atmosphere, and layout. ChatGPT's review, on the other hand, gives a direct, fact-based account that doesn't focus as much on the mood and small details. Then we entered the same prompt for both AI models, as you can see above. For the prompted result, in summary, Claude’s outputs provides a more comprehensive and detailed analysis, focusing on the interplay of old and new elements and the technical aspects of the photography. ChatGPT outputs offer a structured but less detailed explanation, with additional observations about plants and the potential use of the image in artistic contexts. Both analyses capture the atmospheric and visual essence of the image but differ in their depth and scope.

2.Analyzing a Graph

In this part, we will upload the chart of the money supply in the US to analyze with both models. To do that, we will use the chart image and prompt below:

Here is the prompt we will use for analyzing a graph:

What can be taken away from this graph?

Here is the image (or graph) we will use for analyzing a graph:

ChatGPT-4o Outputs:

Claude Sonnet 3.5 Outputs:

Comparison

When we first uploaded the picture of the graph to both AI models without giving them any instructions, they began to analyze it and give a thorough description. But Claude started to tell the observations from that, while ChatGPT only kept it with a description of the graph. After entering the prompt, the conclusion is that the two AI models provide detailed analyses of the M2 money supply graph. Both highlight long-term growth, significant increases post-2008 and around 2020, and inflation concerns. The Claude offers a more comprehensive view, discussing exponential growth, policy impacts, and potential asset bubbles, while the ChatGPT focuses on key trends and the recent plateau. Claude depth provides a richer context, whereas the second is more concise.

WRAP UP

Conclusion

In summary, we’ve tested out two AI models for the visual-related outputs. Both AI models have amazing capabilities, but which one you would like to use depends on preferences since Claude AI is better for detailed, comprehensive analysis. ChatGPT is better for concise, focused analysis in the context of analyzing graphs or vision capabilities.

🛑 Hold on! Before you leave…

Your thoughts mean a lot! Please take a minute to give us your feedback.

What'd you think of today's edition?

Login or Subscribe to participate in polls.

Your insights are invaluable to us. Please share any additional feedback you may have.