Billing
Prepaid credit, spent by the token at each model's listed price. No plan and no subscription.
- Smallest top-up
- $5
- Largest single top-up
- $5,000
- Pay with
- Card or crypto
- Where
- Dashboard, Billing
How a request is charged
- Fresh input tokens are charged at the model's input price, cached input tokens at its cached price, and generated tokens at its output price. The three prices are on Models.
- While a request runs, the most it could cost is held from your credit. When it finishes, is cancelled, or the connection drops, the hold settles to what was actually used.
- A request that fails before a model answers is not charged.
- Any processing fee on a top-up is shown in the total before you pay.
Running low
Turn on the low balance email under Billing and choose the amount it should warn you at, so an agent does not stop in the middle of a job. A request that needs more credit than you have is refused with a 402 that says how much it needed.