Issue - Unable to reach the model provider (GPT-5.1 Codex High)

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

unable to reach the model provider

Steps to Reproduce

in AI Agent Chat

Expected Behavior

should response

Screenshots / Screen Recordings

Image 05-12-25 at 10.18 AM.png
Image 05-12-25 at 10.18 AM.png2507×1583 349 KB

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 2.1.48
VSCode Version: 1.105.1
Commit: ce371ffbf5e240ca47f4b5f3f20efed084991120
Date: 2025-12-04T19:26:27.263Z
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Darwin arm64 25.1.0

For AI issues: which model did you use?

GPT-5.1 Codex High

Additional Information

Error MSG:

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

Error Log:

Request ID: 1adbd029-54e0-4cdf-871d-3b647efe6df8
{“error”:“ERROR_OPENAI”,“details”:{“title”:“Unable to reach the model provider”,“detail”:“We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.”,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:false}
ConnectError: [unavailable] Error
at isu.$endAiConnectTransportReportError (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:6409:408775)
at rMo._doInvokeHandler (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:21873)
at rMo._invokeHandler (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:21615)
at rMo._receiveRequest (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:20377)
at rMo._receiveOneMessage (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:19194)
at zPt.value (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:17286)
at _e._deliver (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:49:2962)
at _e.fire (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:49:3283)
at Lpt.fire (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:6394:12156)
at MessagePort. (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:8973:18439)

Does this stop you from using Cursor

No - Cursor works, but with this issue

2

I am using Codex High model, than also i am facing such issue., when can we expect it tobe resolved?

3

Depends on OpenAI model availability and how many people worldwide now test Codex Max. Not uncommon when new models are released.

Though if the issue is persistent it may have a different cause. Let me know and I will extract it into a separate post again.

4

Following is the error log, i hope that may help you to identify the problem:

Request ID: 4ae16710-3155-4281-8fd6-126521910c1e
{“error”:“ERROR_OPENAI”,“details”:{“title”:“Unable to reach the model provider”,“detail”:“We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.”,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:false}
ConnectError: [unavailable] Error
at isu.$endAiConnectTransportReportError (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:6409:408775)
at rMo._doInvokeHandler (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:21873)
at rMo._invokeHandler (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:21615)
at rMo._receiveRequest (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:20377)
at rMo._receiveOneMessage (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:19194)
at zPt.value (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:7026:17286)
at _e._deliver (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:49:2962)
at _e.fire (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:49:3283)
at Lpt.fire (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:6394:12156)
at MessagePort. (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:8973:18439)

5

@rauf_khatri Could you please try following to determine what causes it?

About connecting to models:

  • Start a clean profile without extensions and settings: in the terminal
    • Windows: cursor --user-data-dir="~/.cursor_temporary_profile"
    • If everything works in the clean profile, it’s definitely an extension or a setting
  • Disable HTTP/2: Settings > search for HTTP/2 > enable Disable HTTP/2
  • Re-login: Settings > General > Log out, then log back in
  • Please share:
    • A screenshot of Settings > Network > Run Diagnostics
    • Logs from Help > Toggle Developer Tools > Console at the moment of the error
    • 2-3 Request IDs from the chat menu (temporarily disable Privacy Mode if it’s enabled)
    • The list of installed extensions

Let me know the clean profile results and attach the diagnostics. If it reproduces in a clean profile on 2.1.48, we will escalate to engineering.

6

ok here’s the response,

  1. start a clean profile: GPT-5.1 Codex high worked in clean profile.
  2. Disable HTTP/2 : chatbox stuck at planning the next move.
  3. Re-login: gave the same “Unable to reach the model provider“ msg in chatbox.
  4. Request ID: aaca8cf1-dc48-4c63-903e-96cbee93d550
    Request ID: 89283025-2f9e-428d-8c35-05a37dfb8c50
    list of installed extensions:
    Python from cursor(dot)com (anyshpere)
    Python from microsoft(dot)com (ms-python)
    Python Debugger (ms-python) from microsoft(dot)com

screenshots attached

Image 05-12-25 at 12.22 PM
Image 05-12-25 at 12.22 PM1128×1604 658 KB

Image 05-12-25 at 12.19 PM
Image 05-12-25 at 12.19 PM808×1307 28.6 KB

7

Thank you for checking the details. From the findings it looks like the issue is your configuration/profile folder. While I can not pinpoint a specific cause on your device, it is usually simpler to clear it and setup fresh as we can not reproduce the issue on our side without much more information.

Could you backup the App Data folders as described in the link below and clear it, then setup the profile fresh with existing app. A basic setup would be sufficient to see if adding the extensions and core settings caused it to go wrong again.