I’m getting “Problem reaching OpenAI” error message. I’ve tried: updating Cursor, closing & reopening, using a different model. I’m not using my API key. Has anyone else had experience with this? What’s the solution?
The request ID is e67330ac-2b1f-404e-a30a-28ad8f520314.
I encountered same issue, on both my local cursor and via ssh server host, for Pro tier user. 
req-id: ed334456-8f6a-470d-9f43-8253df36f19a
i restarted the deployment. can you check again. hoping that it is down now.
Thanks, I realize it may be caused by my network, now I am at office rather than my home, where I need http proxy to access internet, i tried configure http proxy in settings as VS Code do, while it seems not taking effect.
Any suggestions on this, please? Thanks!
Update:
this should be some kind of corruption on configuration level, after cleaning up(with app cleaner on macOS) cursor and a fresh install, everything is fine now.
oh, interesting. i will look into it more! please keep reporting when you see more problems.
Edit 1:
Also another thing , it counts the requests to OpenAI on the dashboard even-though reaching openAI servers failed, If you have a fix please tell me.
Edit 2:
Also another quick observation , it does work for GPT-4 , but gpt3.5 is not working , maybe theres some issue there?
Seems like we we had a server outage at that time but it should be back to normal now. Does this issue still persist for you?
Hey checked just now , seems to be working again , thank you
Yep had an intermittent problem at that time. Apologies about this!
Its 2024 and this is still happening. It happened multiple times this morning
Hey! Could you send over some of the failed request IDs? I’ll take a look!
bbef84de-81de-42bb-a184-9540bbd6a302
bb9c17a6-e060-40dc-b6de-f12c5c2d323f
Spend the day yesterday getting same errors, and this morning as well after it worked for like 10 minutes
Here is another code
97b680bd-405c-411e-a916-ce8fb7a155c9
And another from just today:
3a6d7931-b202-4fce-bdab-4cf25acb7002
It only happens when using Sonnet 3.5 and typically happens after an hour of use. Feels like you’re using OpenAI embeddings and get rate limited when sending context… i don’t know.
Your errors indicate you exceeded the daily rate limit for anthropic with your own api key.
We can certainly be doing a better job of surfacing the correct error message here!
Hi!
I am getting same error. “Problem reaching Open AI”. I tried with both type of keys user and project level, but the error still persists. Please advise if anyone has already found the fix for this issue. Thanks!
