Find answers from the community

Updated 6 months ago

Anyone else getting lots of 500 server

At a glance

The community members are discussing issues with the OpenAIAgent, where they are experiencing frequent 500 server errors. Some community members have noticed this problem, while others have not. The comments suggest that the issue may be related to the OpenAI service itself, as one community member mentioned swapping to GPT 3.5 and having fewer issues. Another community member provided a link to their code, indicating that the problem is not happening 100% of the time, but is more frequent than with the React agent. The community members are trying to determine the cause of the issue and whether there is a reliable way to reproduce it.

Useful resources
Anyone else getting lots of 500 server errors when using an OpenAIAgent?
L
i
7 comments
I haven't noticed this 👀
No other helpful information in the error? I wonder if openai is just having some issues
Or if you have a solid way of reproducing, that would help too
I swapped to gpt 3.5 and it’s been fine, dumb, but fine. I was using Gpt4o, and sometimes, it would just error 500 for a solid min before either fully failing or finishing the process.
I would then switch to react agent and there would be no error, which tells me it’s just the OpenAI agent
I’ll link my current code
https://github.com/idontneedonetho/FrogBot/blob/beta/modules/utils/GPT.py

It’s not happening 100% of the time, but it was more than the react agent hitting max iterations.
Add a reply
Sign up and join the conversation on Discord