CursorAI/Requests

• https://docs.cursor.com/account/usage#fast-and-slow-requests
• There are two types of requests in Cursor,  CursorAI/Slow Requests  and  CursorAI/Fast Requests that have their own pool
• By default, Cursor servers try to give all users fast premium model requests.
• when users run out of fast premium credits, they are moved to a slow pool.
  • Wait times in the slow pool are calculated proportionally to how many slow requests you’ve used
• To bypass wait times entirely, you can enable CursorAI/Usage-Based Pricing pricing (you’ll only be charged for requests beyond your included fast requests)
  • See CursorAI/Settings/Models for an overview of which models are premium and their alternatives.
    • #note All LLM Reasoning Models currently require usage-based pricing