Using Auto - getting a high demand error advising me to switch to Auto!

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

When using Auto, I am getting the following message:

High Load: We’re experiencing high demand for the selected model right now. Please upgrade to Pro, switch to Auto, another model, or try again in a few moments.

Steps to Reproduce

Send any prompt in Auto

Operating System

Windows 10/11

Version Information

Version: 2.6.22 (system setup)
VSCode Version: 1.105.1
Commit: c6285feaba0ad62603f7c22e72f0a170dc8415a0
Date: 2026-03-27T15:59:31.561Z
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

Request ID: 7fa78ec3-67b5-413f-9622-58028ff23bc4
{“error”:“ERROR_RESOURCE_EXHAUSTED”,“details”:{“title”:“High Load”,“detail”:“We’re experiencing high demand for the selected model right now. Please upgrade to Pro, switch to Auto, another model, or try again in a few moments.”,“isRetryable”:true,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
[resource_exhausted] Error
die: [resource_exhausted] Error
at e9A (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34265:23755)
at Z5A (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34265:22658)
at a9A (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34266:6285)
at Xou.run (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34266:10400)
at async qOa.runAgentLoop (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:46830:10206)
at async b5u.streamFromAgentBackend (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:46884:9277)
at async b5u.getAgentStreamResponse (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:46884:13663)
at async vMe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34327:17597)
at async Ea (vscode-file://vscode-app/c:/Program%20Files/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:45811:4826)

Additional Information

If I am using Auto, I shouldn’t be prompted to switch to Auto!

Does this stop you from using Cursor

Yes - Cursor is unusable

5

Hey, thanks for the report. Two things are happening at the same time:

  1. High Load on Auto is a real state. Auto picks a backend model dynamically, and when the whole routed pool hits backpressure, the request can fail with a retryable ERROR_RESOURCE_EXHAUSTED. That’s expected under heavy load. Wait a bit and retry, or pick a specific model.

  2. The error text is wrong, and that’s a known bug. The tip to switch to Auto when you’re already on Auto is misleading, same for the upgrade to Pro tip in some cases. On the agent path, requestedModel doesn’t always get propagated correctly before the message is built. We’re tracking it, but I can’t share an ETA yet.

So if Cursor shows High Load, it doesn’t mean Auto is broken. It means the Auto backend is overloaded right now. Workaround during the incident: retry, or switch to a specific model Sonnet, GPT-5, or Gemini.