Thanks for reporting a bug you have found in Cursor!
Please add the following info to help us diagnose your issue:
I’m not sure what happened, was editing files and rejected a few changes. I get “Unable to reach Anthropic” for any new Composer requests. i show there are 26 files stopped and I cannot seem to find a way to reapply. Save all does not work. I’ve closed out and restarted cursor, same issue. I’m dead in the water, not sure how to get past this.
Request ID: 37a96186-a7d7-485c-b050-5e3c8abc0b83
Using MacOS 15.2, Cursor: Version: 0.45.4
Check the forum to ensure the issue hasn’t been reported already
Provide a clear description of the bug
Explain how to reproduce the bug (if known)
Attach screenshots or recordings (e.g., .jpg
, .png
, .mp4
).
Tell us your operating system and your Cursor version (e.g., Windows, 0.x.x
).
Tell us if the issue stops you from using Cursor.
2 Likes
Hey, can you confirm that the problem persists? Also, what happens if you start a new session?
Started a new chat session and restarted app. Seemed to start working again and have not seen issue since.
I have this too but with Google.
We’re having trouble connecting to Google. This might be temporary - please try again in a moment.
Request ID: 30d54b7d-853f-4d66-9ccc-833c65fa3334
or my own key:
We encountered an issue when using your Google API key. Please check your API key settings and try again. API Error:
{
“error”: {
“code”: 500,
“message”: “An internal error has occurred. Please retry or report in Troubleshooting guide | Gemini API | Google AI for Developers”,
“status”: “INTERNAL”
}
}
I tested all good on openrouter.ai
Same problem from my side, since yesterday. Seems to be a capacity problem, as it has been working all the morning (I am based in Europe).
It is unfortunately common to see this error sometimes when Cursor is at its peak users.
Claude is our most popular model and we are still having issues getting the capacity we need from Anthropic during peak times.
My recommendation if you keep hitting this is to try to use another model for a while. DeepSeek v3 can be enabled in your Cursor settings and has comparatively similar performance to Claude but is not considered a premium model, so has no queues or usage limits.