I am starting to rip my own hair out on this issue. I get the same error message “An internal error has occurred” all the time. It started out appearing randomly, the length of the message doesn’t seem to be the issue. The Token size of the entire conversation is barely 40k tokens. But now the error happens in almost every message.
And it is especially frustrating that there is absolutely no info about this on the documentation, and all I found when researching this issue was a lot of people mentioning a similar problem, and never getting answers either.
Maybe it is a very “on the nose” thing, I would still be happy if someone just pointed it out to me. I can not figure it out by myself.
I am using Firefox, but using other browsers doesn’t change the issue.
I tried changing the models, that doesn’t make a difference.
I tried downloading the chat history and re-uploading it in a new chat, but somehow that doesn’t work either.
I tried creating a completely new conversation and re-inserting all relevant information for my project, but the answers from the model are so vastly different even though the information is the exact same.
I thought that maybe the servers were overloaded, I wanted to change to the billed API plan, but for some reason I can’t get the billing to work. So now I am at a standstill.
If anyone has any idea about what could be causing this, please let me know.
You can switch on 9 different charts to explore your requests, and you can look at the status codeas.(200 ok, 500 internal server error, 429 limit exceed)
In the last 6 hours, I’ve made 99 requests and 6 were successful.
Request no. 7 gave me an 500 server error so I waited and retried. Each time it retried I got another error 500 until I hit the “50 requests per day limit” and I now get 429 errors for hitting the limits.
Grumpy mode:
It’s kind of Google to count requests that fail due to Google’s ownl server errors towards our rate limits!
Why is it whenever Google release new models/functionality it breaks existing models already in production?
Either wait and try again or try switching model as they tend to be broken at different times. I’m using model Gemini-1.5-Pro-001
I keep getting the same issues over the past several days, for saved things I’ve ran many times. It was only erroring if I chose 1.5 Pro 002 but now it does it on anything I choose
Getting the same error almost all the time, Google please do something, I am on the paid plan and getting this instead of prompt responses is not cool!