Implement a X-hour rolling window like Claude Code

Feature request for product/service

AI Models

Describe the request

Instead of being able to depleat all our tokens during a couple of days, it would be better to have the possibility to switch to a rolling window consumption model like Claude Code has. This experience is more consistent throughout the month, and avoid users to run out of tokens, or be switched to pay-per-use too quickly in the month.

1 Like