Max Mode
Max Mode extends the context window to the maximum a model supports, giving it a deeper understanding of your codebase for more accurate edits across large files and complex projects.
How do I enable Max Mode?
Open the model selector in your chat or agent panel and toggle Max Mode on. It's a global setting that stays on across conversations. Some models are Max Mode-only and enable it automatically when selected.
Which models support Max Mode?
Max Mode is available on all models. It's most impactful on models with context windows larger than the default ~200k tokens. Open the model selector to toggle it on for any model.
How does Max Mode affect my usage?
Max Mode uses token-based pricing, so a single request can use significantly more of your usage budget than a normal request.
On current individual plans, Max Mode is billed at the model's API rate. On Teams plans, non-Auto requests include the Cursor Token Rate. On legacy request-based plans, Max Mode adds a 20% surcharge.
If you're watching your usage carefully, use Max Mode selectively for tasks that benefit from the larger context window.
When should I use Max Mode?
Max Mode is for when you want the best possible experience from a model, regardless of cost. It's most impactful on models with context windows larger than the default 200k.
For most coding tasks, the default context window works well.