A: Hello Simar,

Thank you for your comment.

This is a common misconception and we discussed the same issue here a few times. GTP-4 does not need to be aware that it is, in fact, GPT-4. The way language models work is more complicated than that. If you select GPT-4 as the model, we will process your request with GPT-4.

A better way to test the underlying model would be to check the content length it can handle. You can easily verify that GPT-4 can process a prompt of up to 6000 words, which is something GPT-3.5 cannot do due to its 4096-token (~3000 words) limit.

Few other ways of testing it are complex math problems, where you can verify that GPT3.5 falls short.

This prompt is a great example how to compare the output of both models: "Name three celebrities whose first names begin with the `x`-th letter of the alphabet where `x = floor(7^0.5) + 1`"

If you run this a few times with ChatGPT-3.5, you will see that it will give you incorrect answer from time to time and ChatGPT-4 will be correct and consistent at all times.

I hope this addresses your concerns.

Please let me know if you have any other questions.

Best,
Michal