From December, ChatGPT and Codex will have a common cap on add-on, pro and business rates, OpenAI wrote in the help. Quotas are merged because the agent functionality of a job moves to the main ChatGPT. The cost will be calculated in tokens, and it depends on the model and the task: older models and complex tasks cost more. Chat in instant mode does not need a cap. OpenAI expects most of them won't have to wait for a reset more often than they currently do. The total balance and reset time will be displayed in the usage section of the settings. When the quota runs out, you can wait for a reset, upgrade to a higher rate, or buy additional credits for functions where they work.