High-load error tells user to “switch to Auto” while Auto is already selected (misleading copy / state)

Support Bug Reports
,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

When the backend returns ERROR_RESOURCE_EXHAUSTED (“High Load”), the UI message says to “switch to Auto, another model, or try again.” In Agent chat, model is already set to Auto (second dropdown), so the suggestion is incorrect and confusing—looks like a bug in error copy or in how “selected model” is determined for Auto routing.

Steps to Reproduce

Open Agent chat with model Auto selected.
Trigger a request during high load (or whatever causes resource_exhausted).
Observe toast: still recommends switching to Auto.

Expected Behavior

Message should reflect reality, e.g. “High demand—try again shortly” or “try a specific model” only when user picked a fixed model—not “switch to Auto” when Auto is active.

Operating System

Linux

Version Information

Version: 2.4.28
VSCode Version: 1.105.1
Commit: f3f5cec40024283013878b50c4f9be4002e0b580
Date: 2026-02-03T00:56:18.293Z
Build Type: Stable
Release Track: Nightly
Electron: 39.2.7
Chromium: 142.0.7444.235
Node.js: 22.21.1
V8: 14.2.231.21-electron.0
OS: Linux x64 6.8.0-106-generic

For AI issues: which model did you use?

“Auto”

For AI issues: add Request ID with privacy disabled

6df058b6-daee-4870-a888-ad4d008cc35d

Additional Information

Server / API payload (if useful):
{“error”:“ERROR_RESOURCE_EXHAUSTED”,“details”:{“title”:“High Load”,“detail”:“We’re experiencing high demand for the selected model right now. Please switch to Auto, another model, or try again in a few moments.”,“isRetryable”:true,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
Client stack (abbreviated):
[resource_exhausted] Error → streamFromAgentBackend / getAgentStreamResponse / submitChatMaybeAbortCurrent in workbench.desktop.main.js (minified: tpf, Xmf, ngc, OTe).

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

5

Hey, thanks for the detailed report.

This is a confirmed bug. When Auto mode routes to a specific model and that model is under high load, the error message doesn’t reflect that you originally picked Auto. So it ends up telling you to “switch to Auto” even though you’re already using it, exactly like you described.

The team is aware of it. Your report, including the exact payload and repro steps, helps us prioritize it, but there’s no ETA on a fix yet.

Workaround: if you hit this, try selecting a specific model manually instead of Auto. That usually avoids the high load routing issue.

A few related threads if you’re curious:

Let me know if anything else comes up.