Cursor Version:
Version: 1.0.0
VSCode Version: 1.96.2
Commit: 53b99ce608cba35127ae3a050c1738a959750860
Date: 2025-06-04T19:21:22.523Z (14 hrs ago)
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.5.0
My connection:
320Mbps
Latency
Unloaded
9 ms
Loaded
36 ms
Upload
Speed
270 Mbps
Server(s) Madrid, ES
The rest of the week has been working fine.
Here as well ( UK ) - very very slow… ..as in minutes for one request ?
On a fast symmetrical fiber so definitely not the connection.
In fact - this is unusable.
Version: 1.0.0
VSCode Version: 1.96.2
Commit: 53b99ce608cba35127ae3a050c1738a959750860
Date: 2025-06-04T19:21:22.523Z (14 hrs ago)
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 23.5.0
Hey, have you both got fast requests available in your usage, or have usage-based pricing on?
Slow requests are expected if you are in the slow pool!
I’m getting this too. I just updated the app this morning and the requests just get stuck on “generating…” Doesn’t matter which model I try.
same
Yes, always with fast requests. It was like some hours, after that yesterday and today works fine.
Thank you.
Same here, since 0.50. Tried to update on 1.00 and same problem.
So I am waiting on 0.49.6 which works fine.
I still have fast request credits.
Version: 1.0.0
VSCode Version: 1.96.2
Commit: 53b99ce608cba35127ae3a050c1738a959750860
Date: 2025-06-04T19:21:22.523Z (3 days ago)
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.5.0
On the latest version, can one of you post your Console logs by running Developer: Toggle Developer Tools and checking the console tab?
Hello Dan, thanks for your help!
This is what I see:
The error keeps repeating when I send a request in the chat/agent.
Then after 3-4 minutes the answer comes and no other message appears in the console.
I also precise in case that in network diagnostic, this keeps loading (DNS, HTTP, SSL, the rest is ok):
No VPN, no firewall. I tried all HTTP options, without success.
Hey, we believe the error you are seeing could be a red herring.
Can you try rerun Cursor with your extensions disabled? You can do this by running cursor —disable-extensions in a terminal, or by running Reload with Extensions Disabled within Cursor!
Let me know how that goes!
Hello Dan,
Thanks for your help.
I deactivated all the extensions, rebooted cursor, tried several times, but it didn’t help…
The strange thing is that I get a reply, but several minutes after, it’s not fully blocked.
Hey Dan,
I kept digging into the issue and ended up uninstalling a few extensions including the ChatGPT one that connects to my local ChatGPT app on Mac. I’m not sure if that was the root cause, but after removing them, everything started working!
Interestingly, just disabling the extensions wasn’t enough, they had to be removed.
I’ve since reinstalled the ChatGPT extension, and everything still works fine. I can’t say for sure if it was the problem, since I deleted a few others and didn’t keep track of them all.
So, if someone runs into a similar issue, it might help to try uninstalling and reinstalling some extensions.
Thanks again, much appreciated!
Hello,
I have the same problem and basically the same errors in console since the last update I received. I have also tried launching without extensions and the problem persists.
workbench.desktop.main.js:999 [KnowledgeBaseService] Error fetching git upstream URL: Error: No full commit provider registered
at y8i.getGitUpstreamURL (workbench.desktop.main.js:492:31851)
at jJi.w (workbench.desktop.main.js:999:5057)
at jJi.y (workbench.desktop.main.js:999:7893)
w @ workbench.desktop.main.js:999
workbench.desktop.main.js:999 [KnowledgeBaseService] Error fetching git upstream URL: Error: No full commit provider registered
at y8i.getGitUpstreamURL (workbench.desktop.main.js:492:31851)
at jJi.w (workbench.desktop.main.js:999:5057)
at jJi.refresh (workbench.desktop.main.js:999:5446)
w @ workbench.desktop.main.js:999
workbench.desktop.main.js:609 Failed to get Git upstream URL for \home\giuseppe\RemoteDirectory\ccf: Error: spawn /bin/sh ENOENT
S @ workbench.desktop.main.js:609
workbench.desktop.main.js:4021 Error fetching branches: Error: Could not determine default branch
Ie @ workbench.desktop.main.js:4021
workbench.desktop.main.js:6724 Error getting git root Error: Command failed: git rev-parse --show-toplevel
fatal: not a git repository (or any of the parent directories): .git
isRepoReadyForBackgroundComposer @ workbench.desktop.main.js:6724
workbench.desktop.main.js:4021 [Background Composer] No preferred remote found after 10 attempts
_e @ workbench.desktop.main.js:4021
4Uncaught ResizeObserver loop completed with undelivered notifications.
I get no response even after several minutes…
Version: 1.1.3 (user setup)
VSCode Version: 1.96.2
Commit: 979ba33804ac150108481c14e0b5cb970bda3260
Date: 2025-06-15T06:55:04.603Z
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Windows_NT x64 10.0.22631
Same after upgrading to 1.0
Accidentaly gotten one response other than that - endless “generating” despite model or connection protocol settings.
Falling back to 0.5 helps.
Sorry to hear a few more of you are facing these issues!
I’ll throw this to the team today to see what the best next steps are here.
@webxtor If you can, are you able to share the console logs like @dichiara19 did?
You can do this by running the Developer: Toggle Developer Tools command in Cursor, and finding the console tab. Once there, try to submit a prompt and see what (if any) logs you get!
Now it’s happened again, no response in the agent.
Same issue. I tried reloading and executing these commands. It worked on the first response, but it got stuck again on the second one.
Hey guys, thanks for all the info and reports here!
We’ve identified and implemented a potential fix that should be included in our next release.
No ETA on that yet, but has a good chance of improving the performance / issues here!
Every single day for the last week, around 8AM GMT +2 time zone, the responses just become extremely slow. Not able to use it at all. Around 10PM, it starts flying again.
[2025-07-15T07:56:13.524Z] Starting stream
[2025-07-15T07:57:07.947Z] Response: ‘foo’ in 54422ms
[2025-07-15T07:57:09.057Z] Response: ‘foo’ in 1110ms
[2025-07-15T07:57:10.151Z] Response: ‘foo’ in 1094ms
[2025-07-15T07:57:11.172Z] Response: ‘foo’ in 1021ms
[2025-07-15T07:57:12.274Z] Response: ‘foo’ in 1102ms
[2025-07-15T07:57:13.260Z] Result: Error: Streaming responses are being buffered by a proxy in your network environment
[2025-07-15T08:00:25.283Z] Starting stream
[2025-07-15T08:00:25.575Z] Response: ‘foo’ in 275ms
[2025-07-15T08:00:26.539Z] Response: ‘foo’ in 1064ms
[2025-07-15T08:00:27.541Z] Response: ‘foo’ in 1702ms
[2025-07-15T08:00:28.581Z] Response: ‘foo’ in 1740ms
[2025-07-15T08:00:29.555Z] Response: ‘foo’ in 1074ms
[2025-07-15T08:00:30.541Z] Result: true
I’m experiencing significant issues with Cursor after updating to application version 1.1 and since July 3rd: