Where does the bug appear (feature/product)? Cursor IDE Describe the Bug I’ve selected Opus 4.7 and Max mode. The context window tracker shows only 200K instead of 1M. Steps to Reproduce Select Opus 4.7 and Enable max mode. Send a prompt. Hover over the context window tracker icon. Screenshots / Screen Recordings [Screenshot 2026-04-21 at 7.19.37 PM.png] Operating System MacOS Version Information Version: 3.1.17 (Universal) 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: Darwin arm64 25.4.0 Does this stop you from using Cursor No - Cursor works, but with this issue

Where does the bug appear (feature/product)? Cursor IDE Describe the Bug With Max Mode enabled and “1M Max” explicitly selected in the model picker, the context window indicator in the Agent panel consistently shows a 300K token maximum (e.g. ~23.6K / 300K Tokens) instead of the expected 1M tokens. This affects all models that advertise 1M context support — not just Claude Opus 4.6 but also Claude 4.5 Sonnet and others. The 300K cap appears to be a hard UI/backend limit regardless of which 1M-capable model is selected. Per Cursor’s own documentation and pricing page, Max Mode should extend the context window to the model’s full maximum (1M for Opus 4.6 and Sonnet 4.5), with no long-context surcharge. The 1M context window for these models went GA in March 2026. Steps to Reproduce Open Cursor IDE (v3.4.1 Nightly) Open the Agent panel Select any 1M-capable model (e.g. Opus 4.6) Enable Max Mode (toggle “1M Max” in model selector) Observe the context indicator — it shows X / 300K Tokens instead of X / 1M Tokens Repeat with other 1M models (Sonnet 4.5, etc.) — same 300K cap Expected Behavior The context indicator should show X / 1M Tokens (or ~1,000K) when Max Mode is enabled on a model that supports a 1M context window. The full 1M context should be usable before any summarization or condensation kicks in. Operating System MacOS Version Information Version: 3.4.1 VSCode Version: 1.105.1 Commit: 2a298dd06944a9b9ea541d28225b779fcbcc6200 Date: 2026-05-08T16:05:07.818Z Layout: editor Build Type: Stable Release Track: Nightly Electron: 39.8.1 Chromium: 142.0.7444.265 Node.js: 22.22.1 V8: 14.2.231.22-electron.0 OS: Darwin arm64 25.3.0 For AI issues: which model did you use? Claude Opus 4.6 (with 1M Max enabled) — but the issue reproduces with all 1M-capable models. Additional Information This appears related to several other reports: Chat context limited to 70k for a 200K model — same class of bug but with different numbers Code-supernova-1-million model issue — 800K model stalling at 280K (~36%) Context limitation suddenly limits length — premature condensation The pattern suggests Cursor has an internal cap (possibly 300K or lower) that overrides the model’s actual context window, even when Max Mode is toggled on and the UI shows “1M Max” in the model selector. Does this stop you from using Cursor Sometimes - I can sometimes use Cursor

Update: This started happening again today. Yesterday was completely fine, full 1M context working as expected. Same setup, same model, same “1M Max” toggle, now back to ~300K. Honestly, what is Max Mode even for if the full context window isn’t guaranteed? Paying a premium for a feature that silently breaks overnight with no warning is not okay. You only notice when the agent starts losing context mid session because it condensed everything away. Any update on what’s causing these regressions? @Colin @mohitjain

Opus 4.7 1M Max — context window capped at 300K on Ultra (regression) Symptom: Opus 4.7 1M Max shows ~XXK / 300K Tokens in the Context panel instead of the expected 1M denominator. Reproduces on every chat (new and existing), across full Cursor restarts, with Max Mode toggled off->on, etc. Worked correctly: A few days ago. Broken since: ~May 9–12, 2026 (correlates with build 3.3.30). Environment: Cursor: 3.3.30, build date 2026-05-09T18:28:42Z, commit 3dc5592 Plan: Ultra ($200/mo), on-demand at $1329.57 / $2000 Model: claude-opus-4-7-thinking-max (“Opus 4.7 ⊕ 1M Max” in selector) Max Mode: ON (verified in model selector) OS: Windows 10.0.26200 Request ID: What I’ve ruled out: Max Mode toggle is on (screenshot attached). Model slug is the 1M Max variant (screenshot attached). Plan supports it (Ultra includes Max; on-demand has $670 remaining). Full Cursor restart doesn’t clear it. Fresh chats also show 300K denominator. No relevant settings.json customization. Reference — adjacent Opus 4.7 1M regressions in the same week: anthropics/claude-code#55504 (Opus 4.7 [1m] capped at 200K) anthropics/claude-code#54056 (auto-compact at ~367K) anthropics/claude-code#53199 (auto-compact far below 1M) Different products, but the timing and shape of the regression suggest a shared upstream issue with Opus 4.7’s 1M handling. Happy to provide more diagnostic info on request.

Same here. I saw opus 4.7 1M MAX the first time I enabled max mode (matching the GPT-5.5 1M median UI), but after a quick toggle, it only displays opus 4.7 MAX. I still haven’t been able to get the 1M context for Opus 4.7 to work consistently. [image] [image] [image]

[image] Same issue is happening with me, yesterday it was working well but today any model I choose with 1M it defaults back to 272K tokens and starts chat summarizing. Please help

So it’s not just me. Multiple people reporting the same regression today, all saying yesterday was fine. Looks like something broke on the backend overnight. @Colin @mohitjain @Condor @Sanjeed5 can we get some eyes on this?

Guys, Cursor is supposed to be industry standard, the bugs are getting more and more frustrating. This is honestly embarrassing.

Same issue here and I can clearly see that we are billed the MAX mode price although it is broken

7 people in this thread alone, all reporting the same thing today. No staff response yet. Can someone from the team please acknowledge this?

Opus 4.7 Max mode shows only 200K context instead of 1M

Support Bug Reports

Georgewize (George) May 12, 2026, 5:07pm 34

@Yasser_Elsayed @Kazopl Look here Context decrease - Support / Bug Reports - Cursor - Community Forum

Topic		Replies	Views
Context decrease Bug Reports max-mode , context , anthropic	2	121	May 14, 2026
I selected the Opus 4.7 1M model, but the actual context limit is 300K Bug Reports max-mode , context , anthropic	1	238	May 12, 2026
Opus 4.6 Max stuck at 100% context even in brand-new chats Bug Reports max-mode , context , anthropic	6	181	April 30, 2026
Max Mode shows 300K context limit instead of 1M for all models (Opus 4.6, Sonnet 4.5, etc.) Bug Reports nightly , max-mode , context , anthropic	1	853	May 12, 2026
1M doesnt change context window Bug Reports max-mode , context , openai , anthropic	3	268	May 4, 2026

Opus 4.7 Max mode shows only 200K context instead of 1M

Related topics