Balance and Quotas

RouteAPI is prepaid: you top up your account first, and each call deducts the actual usage from your balance. You can set a quota for each API key individually, or enable auto top-up to avoid service interruptions when the balance runs out.

Priced in US dollars

Account balance, model prices, API key quotas, and top-up amounts are all denominated in US dollars (USD). The console may convert amounts into another currency for display according to site settings; this is display-only, and charges and accounting are always in USD.

Account balance

The account balance is the shared pool for all your API keys. View the current balance and historical usage on the Wallet page.

• Every request is charged from the account balance, and the balance never goes negative.
• If the balance cannot cover the estimated cost of a request, the request is rejected with 403 and the error code insufficient_user_quota. See Billing and Rate Limits for how the estimate works.
• If your account has an exclusive discount, the Model marketplace shows the discounted price, the usage log marks it as an “exclusive rate”, and the charge is calculated at the discounted price.

API key quotas

When creating or editing a key on the API Keys page, you can set the following for each key:

Setting	Description
Quota (USD)	The maximum amount this key can spend; can also be set to unlimited
Expiry	The key stops working after expiry; defaults to never expiring
Model restrictions	Only allow this key to call the specified models
IP allowlist	Only accept requests from the specified IPs

A key quota is an upper limit, not a separate wallet. A request deducts from both the key quota and the account balance; if either is insufficient, the request is rejected. Even with an unlimited quota, the amount actually available is still limited by the account balance.

A common practice is to create separate keys for different projects, environments, or team members, each with its own quota, so that a single runaway application cannot drain the whole account. The API Keys list shows the used and total quota for each key.

Situation	Response
Key quota exhausted	401, indicating the key quota is exhausted
Remaining key quota cannot cover this request’s estimated cost	403, error code pre_consume_token_quota_failed
Key expired or disabled	401
Requested a model the key does not allow, or the source IP is not in the allowlist	403

Top-up

On the Wallet page, choose an amount and a payment method to top up; the minimum is 1 USD. The available payment methods are those shown on the wallet page.

• The balance is credited with the USD amount you enter.
• When paying by card through Stripe, the actual charge includes a 4% fee on the top-up amount, covering payment processing fees and taxes. The page shows the actual charge before you pay.

Auto top-up

After enabling auto top-up, when the balance falls below the threshold you set, the system automatically tops up a fixed amount using your saved card. This suits production environments that need to run reliably over the long term. Configure it under “Auto top-up” on the wallet page.

Setting	Description	Range
Trigger threshold	Triggers a top-up when the balance falls below this amount	At most 1,000 USD, default 20 USD
Top-up amount	Amount for each auto top-up	10–5,000 USD, default 100 USD
Monthly cap	Cumulative cap for auto top-ups within a calendar month; 0 means unlimited	At most 50,000 USD, and not less than the top-up amount
Payment methods	Cards used for charging	Up to 3 saved, used in the configured order

How it works:

• The balance is checked after each request settles, and a top-up is triggered if it is below the threshold. Two auto top-ups are at least 10 minutes apart.
• If the balance is already below the threshold when you save the settings, a top-up is triggered immediately.
• If the default card is explicitly declined (for example, the card has expired or the issuer declined it), the next saved card is tried in order. When the outcome is unclear (for example, 3D verification is required), the card is not switched, to avoid double charges.
• After all cards fail, retries happen automatically after 15 minutes, 1 hour, 3 hours, and 6 hours. If they still fail, retries stop and you are notified by email. The wallet page also shows a prompt; after changing the payment method, you can click “Top up now”.
• Once the monthly cap is reached, no further auto top-ups occur that month.
• Auto top-ups are also subject to the 4% fee.

Tip

Keep the trigger threshold below the top-up amount and leave headroom for your peak burn rate. If you might spend the threshold amount within 10 minutes, raise the threshold or the top-up amount.

Balance alerts

In the notification settings of Personal settings, you can set a balance alert threshold; when the balance falls below it, you are notified by email, Webhook, Bark, or Gotify. The system also sends an email reminder when requests are rejected because the balance is exhausted, at most once per day.

What to do when the balance is insufficient

1. Top up on the wallet page, or enable auto top-up.
2. If the error code is pre_consume_token_quota_failed, or the message says the key quota is exhausted, raise the quota for that key; topping up the account will not help.
3. If the balance looks sufficient but requests are still rejected, check whether max_tokens in the request is set too high. The estimated cost includes max_tokens; see Billing and Rate Limits for details.
4. Check recent consumption in the usage log.