Cursor not working!

[babyshark8386 (Dat)]

Describe the Bug

I cannot chat using Chat GPT 5.2

Steps to Reproduce

I write command, and select model, it’s not working. It’s say:

We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.

Screenshots / Screen Recordings

Operating System

MacOS

Version Information

Version: 2.4.21
VSCode Version: 1.105.1
Commit: dc8361355d709f306d5159635a677a571b277bc0
Date: 2026-01-22T16:57:59.675Z
Build Type: Stable
Release Track: Default
Electron: 39.2.7
Chromium: 142.0.7444.235
Node.js: 22.21.1
V8: 14.2.231.21-electron.0
OS: Darwin arm64 25.2.0

For AI issues: which model did you use?

Chat GPT 5.2

For AI issues: add Request ID with privacy disabled

Request ID: 9035d406-dbe7-4c58-9fa2-1643c354925e
{“error”:57,“details”:{“title”:“Provider Error”,“detail”:“We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.”,“isRetryable”:true,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
[resource_exhausted] Error
LTe: [resource_exhausted] Error
at kmf (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:9095:38337)
at Cmf (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:9095:37240)
at $mf (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:9096:4395)
at ova.run (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:9096:8170)
at async qyt.resume (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:34190:60450)
at async Wpc.streamFromAgentBackend (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:34239:7695)
at async Wpc.getAgentStreamResponse (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:34239:8436)
at async FTe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:9170:14575)

Does this stop you from using Cursor

No - Cursor works, but with this issue

[user1717 (Two One2022)]

me too

Version: 2.4.21 (system setup)
VSCode Version: 1.105.1
Commit: dc8361355d709f306d5159635a677a571b277bc0
Date: 2026-01-22T16:57:59.675Z
Build Type: Stable
Release Track: Default
Electron: 39.2.7
Chromium: 142.0.7444.235
Node.js: 22.21.1
V8: 14.2.231.21-electron.0
OS: Windows_NT x64 10.0.26100

Request ID: 981700e9-0b8e-4251-a59a-cebd84667a1f
[resource_exhausted] Error
LTe: [resource_exhausted] Error
at kmf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9095:38337)
at Cmf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9095:37240)
at $mf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9096:4395)
at ova.run (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9096:8170)
at async qyt.runAgentLoop (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34190:57047)
at async Wpc.streamFromAgentBackend (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34239:7695)
at async Wpc.getAgentStreamResponse (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34239:8436)
at async FTe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9170:14575)
at async Oi (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:32991:3808)

[deanrie]

Hey, thanks for the report.

For @Andrii_Vedmid this is a known issue with the SSL handshake over HTTP/2. Try this workaround:

1. Cursor Settings → search for “HTTP/2” → set HTTP Compatibility Mode to HTTP/1.1
2. Restart Cursor

More details in the consolidated thread: [Known Issue] HTTP/2 Connection Failures with SSLV3_ALERT_HANDSHAKE_FAILURE

For @babyshark8386 and @user1717 you’re getting a resource_exhausted error. Try the same HTTP/2 workaround, and also:

• Go to Settings → Network → Run Diagnostics (please share a screenshot of the results)
• Confirm if you’re using a VPN, proxy, or firewall (for example, Zscaler)
• What region/country are you in?

Let me know if disabling HTTP/2 helps or not.

[user1717 (Two One2022)]

i can use Auto mode and third party model provider that i added.
other models that cursor provides are unusable

[deanrie]

Thanks for the screenshots, that’s an interesting situation. Usually HTTP/2 fails the diagnostics, but in your case it’s the opposite, HTTP/1.1 shows a ping issue.

A few questions:

1. Have you tried switching to HTTP/1.1 and restarting Cursor? Even if the ping test is red, models can still work, try sending a message in chat.
2. What region are you in? Are you using a VPN, proxy, or a corporate firewall (like Zscaler)?
3. If you’re using a VPN, try both options (HTTP/2 and HTTP/1.1) without it.

There’s a known HTTP/2 bug for some regions (mostly Ukraine).

The fact that Auto mode and third-party models work suggests the issue is specifically with the connection to Cursor’s providers, not Cursor itself.

Let me know the results and we’ll figure it out.

[David_Ferreira]

I’m having the same problem.

[user1717 (Two One2022)]

LOL,it turned out it’s just because i’ve set my key to use the third party GLM-4.7. i toggle it off and everything is good for me now

[user1717 (Two One2022)]

i know it might affect my use of the GPTs, but i didn’t expect that it will stop me from using opus

[David_Ferreira]

How can you disbale the third party GLM-4.7. if we aren’t using external models?

In my case I’m trying to use opus 4.5 thinking and by the looks @babyshark8386 was using GPT-5.2.

[user1717 (Two One2022)]

maybe you are using your own key?idk,but i think cursor can do a better job at inform people about the behavior of the custom key setting.I have no idea why a openai key will stop me from using opus.