What do overages mean?

We track input and output tokens separately. That means even if you’re staying within your total token limit per minute, it’s still possible to go over on just input or just output tokens.

Why does this matter?

Right now, we don’t charge for overages—but we might in the future. If that happens, you'd only be billed for the part you went over on.

For example:

Let’s say your limit is 600,000 tokens per minute.

• You use 599,000 input tokens

• You also use 61,000 output tokens

That adds up to 660,000 total tokens. While you’re under the input cap, you’ve gone 1,000 tokens over on output—and that’s what could be billed as an overage in the future.

What should I do now?

There’s no action needed at the moment. We’re not billing for overages today. The overage graphs you see are just there to help you track how your input and output tokens are being used.

Keeping an eye on these can help you stay within limits if/when billing for overages kicks in.