What is going on with the code block output of ChatGPT the last few days? It keeps breaking every now and then since 2024-01-13.
It’s like it gets a random linebreak inserted which breaks its flow of code and it exits the codeblock, and then when it meant to exit, it enters a new one.
Not only is this stressful and hateful but it brings with it so many problems.
The first of all is that of memory: chat does not follow instructions, not even those of the prompt itself.
It doesn’t remember anything and I have evidence to prove it, where I ask him to write me the class for a GUI with the name “GUI” and he writes me a class “MainClass”.
The line breaks seems ti be the last of the problem.
I would like to know how to contact OpenAI to ask them to resolve this problem because €20 for this service is INDECENT. Or at least I would like them to read our forum chats to try and help us. Its frustrating.
Especially since you can’t have a 40 mesagges limit and have to use half of them to regenerate responses due to network errors.
There are THREE problems:
1 - wrong format
2 - network errors
3 - memory and fidelity to instructions.
In January it worked like a charm, so much so that in a week I made an entire project in Java.
Now it’s like talking to a child with ADHD who hears but doesn’t listen. Please OpenAI try at least to fix this.
Yes, i got booted fro GPT-4 because it made about 20 errors in a row, Regenerate, regenerate, server problem, something else… all within the space of 20 minutes and then i got booted for 2 hours after that.
I am in the middle of working and this really costs a lot when you are depending on it to be accessible.
This code block format broken issue is really annoying! Especially when using GPT4… When using Bard, there is no problem. But Chatgpt UI dev team is not fixing this critical issue. Why?