Week 2 Exercise - Head-to-Head: Evaluating AI Models

Hello ​@parwalrahul ,

Both models are strong in their own ways. ChatGPT 4.0 excels at generating well-structured, in-depth responses that feel natural and insightful. Gemini, on the other hand, focuses more on straightforward, factual delivery. If the task requires detailed analysis or a human-like tone, ChatGPT 4.0 is the better choice. If brevity and direct accuracy are the priority, Gemini performs well.

Ultimately, the best model depends on the specific use case.

Thanks,

Ramanan 

Hello

ChatGPT 4.0 and Gemini make a powerful impression but have different strengths. 
ChatGPT 4.0 offers strong text-based capabilities and delivers creative results. Sometimes somewhat biased answers. You have to be able to read and evaluate the answers correctly. Provides answers even when ChatGPT doesn't know - creative.
Gemini has a multimodal design and the integration with Google services provides a versatile and comprehensive user experience. 
I think the choice between the two models depends on the specific user needs, such as the need for multimodal processing, integration with existing tools or the focus on unbiased answers.

Frank

​@Ramanan nice takeaway.. my own takeaway is pretty similar to that.

One clear area where gemini excels is image recognition (OCR), especially if we go into native (local) languages… Google really has some edge there.

​@Frank Kokoska yeah, for people in the google ecosystem or services, Gemini could really stand out.

I feel similar possibilities will become a reality with copilot (backed by ChatGPT). already Microsoft is integrating it with office applications.

Interesting times ahead 🤞

Hello,

So both Chat GPT 4.0 and Gemini are good and powerful in their own ways. However I found Gemini provides more precise and detailed information when it comes to technical details on application, tools, languages.

Can we get the recordings or minutes of meeting for  week2 session?

​@KajalS “technical details on application, tools, languages.”

what specific differences did you noticed? interested to know more about this.

Sure Rahul. One of my question was around how can I integrate API tests written in postman with a pipeline on circleCI.

Gemini responded with almost similar steps but along with few examples with detailing like -

1. .circleci/config.yml  looks like and what each keyword specifies with in the file (for e.g. -e environment.json,  --reporter cli,juni).
2. Tips like - Important Security Considerations:
3. Format of environment.json file to store environment variables.

My question was also around Cypress tool, how to start with it.  It replied with basic information like how to install it, key cypress commands, assertions and examples.

Where as Chat GPT provides more theoretical answers(unless specified more accurately) like key features, advantages, etc.

Both ChatGPT and Gemini are powerful AI tools that provide in-depth insights, though they take different approaches.

• ChatGPT offers a well-rounded perspective by highlighting free AI tools for accessibility testing, detailing how each tool targets specific functional areas. It also outlines key components of accessibility testing, covering both manual and automated approaches.

• Gemini focuses on the role of AI in accessibility testing, explaining how AI is integrated into the process. It provides a detailed analysis of the advantages and disadvantages of using AI for accessibility testing.

Conclusion:
Both models deliver comprehensive information, but the choice depends on specific needs. ChatGPT is ideal for those looking for practical tools and a balanced approach, while Gemini is better suited for exploring AI-driven accessibility testing in depth.

​@Bharat2609 nice try.

Multimodal processing is indeed an amazing possibility and would be the thing that would get more prominence.

We might be finetuning the responses from public LLMs through our custom models and vice versa.

thanks for sharing this possibility. Have a nice day!