Unexpected MAX Usage Enablement

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

I never enabled MAX mode in the 400+ days since my accounts inception.

I opened a new chat and it was Grok 4.5 High MAX mode. I never enabled my account for MAX mode. Then I switched to Codex 5.3 Extra High, which was also in MAX mode by default. I have never ever used MAX mode so I sent the request through excepting it to stop and say “Enable MAX” in the dashboard. The request on MAX mode went through which I didn’t expect and it consumed “2.2” requests (5.3 x high is one request and xhigh fast is 2 requests per prompt) before I stopped it.

Steps to Reproduce

I cannot reproduce this because my new chats open with my preferred 5.3 Codex Extra High as expected

Operating System

MacOS

Version Information

Version: 3.9.16
VS Code Extension API: 1.105.1
Commit: 042b3c1a4c53f2c3808067f519fbfc67b72cad80
Date: 2026-06-27T06:41:01.941Z
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 40.10.3
Chromium: 144.0.7559.236
Node.js: 24.15.0
V8: 14.4.258.32-electron.0
xterm.js: 6.1.0-beta.256
OS: Darwin arm64 23.6.0

Does this stop you from using Cursor

No - Cursor works, but with this issue

Hey, thanks for the report. I checked your account and there isn’t any extra charge, and here’s why.

The Enable MAX in dashboard gate you expected is behavior from legacy request-based plans. On usage-based Pro, the MAX request just goes through without a separate confirmation. Also, the 20% MAX upcharge was removed a few months ago, so even if the picker showed MAX, there’s no extra fee for it.

I checked your usage. The requests you sent were counted as included, with no request-based surcharge. The 2.2 requests is the normal token-based calculation for a heavy Codex 5.3 Extra High prompt xhigh plus fast turn, not a MAX penalty.

The fact that a new chat opened with a MAX badge is a transient display quirk in the model picker and it’s a known issue on our side. Based on what you described, new chats are already opening with your preferred Codex 5.3 Extra High.

For the future, for usage-based pricing we’re removing MAX Mode from the model picker on July 20. The same context window sizes will still be available, you’ll just pick them directly in the picker without a separate MAX toggle. Full team write-up here: What does Cursor removing “Max Mode” mean for usage-based pricing users?

If you see a real extra charge for MAX or something doesn’t match in usage, send the Request ID and I’ll take a look.