Auto model is slow [again]

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Cursor is stuck on “Taking longer than expect” for a long time

Steps to Reproduce

Just run any request with auto model

Expected Behavior

Normal speed

Operating System

Windows 10/11

Version Information

Version: 3.1.17 (system setup)
VSCode Version: 1.105.1
Commit: fce1e9ab7844f9ea35793da01e634aa7e50bce90
Date: 2026-04-19T19:33:58.189Z
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 39.8.1
Chromium: 142.0.7444.265
Node.js: 22.22.1
V8: 14.2.231.22-electron.0
OS: Windows_NT x64 10.0.26200

For AI issues: which model did you use?

“Auto”

For AI issues: add Request ID with privacy disabled

c79150d1-2b0c-4c82-89ae-04187618b660

Additional Information

None

Does this stop you from using Cursor

Yes - Cursor is unusable

It’s seems that the model is back again, speed is OK now. I do not understand why we are getting so much issues with speed the last couple of days.

Hey, thanks for coming back with an update, glad it cleared up.

About the repeated slowdowns, Auto routes between multiple models, and sometimes specific providers or routes on our side start lagging, which causes the Taking longer than expected message. It usually fixes itself after a few minutes, like you saw.

If it happens again, send the Request ID here plus an approximate timestamp with your timezone when it was really bad. That helps a lot to match it with our backend metrics. As a temporary workaround, you can manually switch to a specific model Sonnet or GPT-5, it often helps bypass a stuck route.