Unable to reach OpenAI. (but I am using Claude 3.7 Thinking)

Bug Reports
1

I have been using Claude 3.7 Thinking and as it was finalizing a requested change, it just stopped and gave the following:

Unable to reach OpenAI. (try again) Provider was unable to process your request.

Try again does not do anything and I have not been able to use Cursor. It was in the middle of applying changes so changing models is not optimal.

MacOS version 15.3.1 (24D70)
Cursor Version: 0.46.8
VSCode Version: 1.96.2

:prohibited: this issue stopped me from using Cursor.

1 Like
2

Currently having the same issue. I restarted cursor, it then worked for 2 prompts, then continued back to ā€˜Unable to reach openaiā€™ (im using Claude 3.7 thinking - agent). I restarted again, and still the same issue for the past 10 minutes. :frowning:

1 Like
3

Iā€™m having the same issue. My work around is to quit and restart cursor and it works most of the time. But if I send more than one or two prompts, problem returns til restart.

4

Experiencing the exact same issue. using Cursor latest update 0.46.11

RequestID: 8036d088-ba1e-402a-8db0-ac6f627563f7

Unable to reach OpenAI

5

Experiencing exactly the same issue. Have not found a workaround.

Request ID: a1fa98c2-f267-4539-932a-1b3d853e24dd

6

Ah, I had ~12 MCP connections, and decided to disable most, so now Iā€™ve just got 5 enabled. The Agent began working for me after doing this. May have been MCP server tool context overload.

7

For anyone having this issue, just starting a new chat worked for me. Also restarting sometimes did. Not sure if it will return but trying that worked for me!

8

the same issue,This situation has been going on for two days.
IDļ¼š8ee9ed7b-5a90-4067-b1e2-ffb76de57447

9

Hi all that are having the same issueā€¦

I was able to resolve this issue simply by starting a NEW (+) chat after i restarted Cursor. I copied the chat messages that were relevant or that would provide context for the new chat session in my notes, since the AI Agent would be starting a fresh conversation with no context or background on what was being worked on. I make a habit of telling the AI Agent to explain the project and what it is working on now, completed and any upcoming tasks based on its understanding to provide the background and reminder it may need later as a refresher. This helps a lot and will save a lot of time when your agent starts exhibiting dementia and tells you it cannot do tasks that it had been doing all day (which usually means you need to restart and start a new chat). Good luck and hope this helps.