Hi,
I have a couple of questions about the Business plan and how it works in practice:
When an individual user exceeds the 500 request limit, I noticed that the service does not seem to slow down. Is this limit calculated as an average across team members? I also see the metric “Premium Requests Used (user/mo)” — is this related to how limits are applied?
Additionally, I sometimes reuse or rewrite prompts. Does the previous prompt remain in the context and potentially affect accuracy? I’m concerned that if the previous prompt is unrelated, it could consume context space and make responses less accurate.
Any insights or clarification would be appreciated!
Thanks in advance.
For Business plans, request limits aren’t averaged across team members - each person gets their own individual 500 request quota that resets monthly. You can check out the details at Cursor – Plans & Usage
The “Premium Requests Used” metric shows how many premium model requests you’ve used out of your personal monthly allocation. If you’re not seeing slowdowns after 500, you might have usage-based pricing enabled which keeps things running smoothly
As for prompt reuse, previous messages do stay in the context, and remnants of the files previously read in your current chat will always be fed into any following prompts. We try to manage this for you so you’ll never see an error, but we do often recommend regularly starting new chats (CMD+N) just to keep things clear!