Where does the bug appear (feature/product)?
Cursor IDE
Describe the Bug
I’m on a Cursor Pro individual plan. For the past week, using Claude Opus 4.8 (all configurations — thinking/non-thinking, high/xhigh, fast/standard) intermittently worked without Max Mode enabled. I could toggle Max Mode off, send a request, and it processed normally at 1-2 requests per prompt.
As of today (08/07/2026), toggling Max Mode off and sending a request now triggers a backend error above the chat input:
Max Mode Required
The model “claude-opus-4-8-thinking-high” requires Max Mode to be enabled.
Please enable Max Mode and try again.
The UI toggle itself still works (Max Mode can be switched off), but the server now rejects any Opus 4.8 request sent without it. This is a server-side enforcement change — the client UI hasn’t changed.
When Max Mode is ON, requests process but bill through the API pool at per-token rates (~20x the cost vs. the 1-2 fixed requests it consumed last week).
There is no mention of this in the Cursor changelog(or did i miss something?).
Steps to Reproduce
- Open Cursor IDE
- Open Agent or Chat panel
- Select Opus 4.8 (any config: claude-opus-4-8-high, claude-opus-4-8-thinking-high, claude-opus-4-8-fast)
- Toggle Max Mode OFF (the toggle works and allows being turned off)
- Submit a prompt
- Observe error: “Max Mode Required — The model requires Max Mode to be enabled.”
- Toggle Max Mode ON → request processes but bills at API-pool per-token rates, not 1 fixed request
Expected Behavior
Based on last week’s behavior on the same plan: Opus 4.8 should process with Max Mode off at N fixed request per prompt — or if always requiring Max Mode is intentional, this should have been announced.
Screenshots / Screen Recordings
Operating System
Windows 10/11
Version Information
Version: 3.10.20 (user setup)
VS Code Extension API: 1.125.0
Commit: 23b9fb205fe595ea2be29da7214e19762d037fc0
Date: 2026-07-07T07:03:33.071Z
Layout: IDE
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: Windows_NT x64 10.0.26200
For AI issues: which model did you use?
All latest and frontier models
Additional Information
- I am on a Cursor Pro individual plan (not Team/Enterprise)
- I understand Opus 4.8 draws from the API pool, but the sudden server-side rejection — from working intermittently last week to fully blocked today, no changelog — suggests a backend enforcement change deployed without notice
- Cursor’s March 16 FAQ: “Are individual plans affected? No, with the exception of GPT 5.4”
- Similar backend regression occurred April 28-29, 2026 (“Max Mode default on all GPT legacy models”) where staff confirmed incorrect routing and later fixed it
Does this stop you from using Cursor
Sometimes - I can sometimes use Cursor