Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I’m surprised you are able to get a whole script working as expected out of it. I’ve tried using chatgpt just for lines of code alone and its always coming up with a solution thats, I guess you could say is far to “creative” to be useful and often doesn’t end up doing what I expect of it when I go to test what its given me.


I had a very specific ask. I gave it the uri as comma separated values and asked for the swagger spec. There wasn’t scope for creativity.

I could also split the uri by service names. That helped parallelize my questions. It wasn’t just dump the data in. There was some cleanup behind the scenes that I had to do.


GPT-4 or GPT-3.5?

They're completely different products.


We need to stop propagating this nonsense. GPT-4 still messes up.


OP said they're different products, which is very true. Yes, GPT-4 still makes mistakes, but it's leagues ahead of 3.5.


It's not nonsense at all. The degree to which GPT-4 messes up is dramatically lower than 3.5.


The nonsense is the automatic conclusion that when ChatGPT messes up, it must be 3.x; which is demonstrably false.


But no one said that. GP just said they were different products. And that is true. It is meaningless to show an error that GPT 3.5 produced in a conversation about 4. They are separate products.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: