Cursor’s context window depends on the model you pick: Grok 4.7, Cursor’s flagship model, uses a 256k-token window by default and can go up to 500k, while Composer 2.5 is listed at 200k, according to Cursor’s documentation.
As of November 9, 2026, Cursor lists larger maximums for third-party models: Claude Opus 5.5 starts at 300k tokens, GPT-5.6 Sol at 272k and Gemini 3.1 Pro at 200k, and each can go up to 1M. Third-party models require the Premium plan or higher. On the Plus plan, Grok 4.7 is fixed at its 256k window.
Larger windows cost more. Grok 4.7 requests with more than 256k input tokens bill at twice the standard rates, or three times in Fast mode, and GPT-5.6 long context doubles the input price. Max Mode, which used to unlock bigger windows, now applies only to legacy request-based plans. A Cursor staff reply on the Cursor forum says you now hover a model in the chat model picker, click Edit and set Context to 1M.
Related: Cursor usage limits on each plan, How much does Cursor Pro cost? and What is the Claude context window?
Sources: Cursor Docs, Models table; Cursor Docs, Models & Pricing; Cursor Forum, staff reply on 1M context
Image: Cursor logo by Anysphere, public domain, via Wikimedia Commons. Placed on a white background and resized. Cursor is a trademark of Anysphere. FreeAnswers is not affiliated with or endorsed by Anysphere.