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

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Request ID: d9cd2cd7-17bf-4d24-94e0-420a40ebb790
Provider Error We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.
H4t: Provider Error We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.
at lpf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9097:38263)
at apf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9097:37251)
at Cpf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9098:5088)
at Tva.run (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9098:9098)
at async Vyt.runAgentLoop (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34206:57104)
at async mgc.streamFromAgentBackend (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34255:7835)
at async mgc.getAgentStreamResponse (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34255:8576)
at async qTe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9180:14965)
at async Xi (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:33004:3808)

Steps to Reproduce

ask any AI

Operating System

Windows 10/11

Version Information

Version: 2.4.32 (system setup)
VSCode Version: 1.105.1
Commit: dac77182205e080f83e43f13ee42d822740a6e10
Date: 2026-02-09T05:01:38.814Z
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 10.0.26200

For AI issues: which model did you use?

Composer 1.5
GPT-5.3 Codex
GPT-5.2

For AI issues: add Request ID with privacy disabled

Request ID: d9cd2cd7-17bf-4d24-94e0-420a40ebb790
Provider Error We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.
H4t: Provider Error We’re having trouble connecting to the model provider. This might be temporary - please try again in a moment.
at lpf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9097:38263)
at apf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9097:37251)
at Cpf (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9098:5088)
at Tva.run (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9098:9098)
at async Vyt.runAgentLoop (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34206:57104)
at async mgc.streamFromAgentBackend (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34255:7835)
at async mgc.getAgentStreamResponse (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34255:8576)
at async qTe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9180:14965)
at async Xi (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:33004:3808)

Does this stop you from using Cursor

Yes - Cursor is unusable

Hey, this is a known issue. We saw the same thing with a few users last week.

Most of the time it’s caused by MCP servers or third-party API key settings that block the connection. Try this:

  1. Cursor Settings > Tools & MCP: If you have any MCP servers connected (especially Perplexity, Figma, or custom ones), temporarily disable all of them.

  2. Cursor Settings > Models: Make sure “Override OpenAI Base URL” is turned off. If you added any third-party API keys (OpenAI or Anthropic), temporarily disable those too.

  3. Sign out and sign back in (Cursor Settings > Sign Out, then Sign In).

  4. Open a new chat in a different project folder to test.

This worked for other users with the exact same issue:

If that doesn’t help, try launching Cursor from Terminal with cursor --disable-extensions to rule out extension conflicts.

Let me know how it goes.

1 Like

Hi Dean,

Yes, disabling the perplexity MCP did the trick.

Best,

Pablo Hortal
CEO
Pix Videos Production Company
Phone (Europe): +34 606357902
Phone (US): +1 (713) 574-1180
[email protected]
www.pix-videos.com

1 Like

Hi Dean, thanks for your advice. Disabling MCP was the solution

1 Like

hmm.. i have the same issue - i am a fresh user with only the basic install going on ( using local Cursor though ) for coding, with Sonnet 4.5 as the engine.
I had so many issues, first it started redoing itself, back and forth in an endlessloop, and after i forcibly stopped it, this Connection issue started - only to end up in me using my montly limit .. in 2 days …
I love what Cursor is doing for me , but i dont much enjoy that i probably lost 25% of my limit in issues …

I don’t have any MCP tools and the Override OpenAI Base URL was toggled off, but I still get this

what else to try?

I switched models, solved , no further action

I just tried one thing - i switched from HTTPS/2 to HTTPS/1.1 - and so far its working … ( who wanna bet im gonna jinx it now ? )

.. Settings - NETWORK - Http Compability mode ..

edit : 4 hours later, still working like a charm …

Glad to hear both of you found workarounds.

@Vidar_Halvorsen switching to HTTP/1.1 is a solid fix. If you run into this, go to Settings > Network > HTTP Compatibility Mode and switch it to HTTPS/1.1.

@Chris_Sleat glad switching models helped. If it happens again, try the HTTP/1.1 trick above too.

For anyone finding this thread later, here are the main fixes that worked:

  1. Disable MCP servers, especially Perplexity, Figma, and custom ones, in Cursor: Settings > Tools & MCP
  2. Turn off “Override OpenAI Base URL” and remove any third party API keys
  3. Switch HTTP Compatibility Mode to HTTPS/1.1 in Settings > Network
  4. Sign out and sign back in, then test in a new chat