Skip to main content
Your plan already caps requests per minute, per day and per month across the whole account. Quotas are the layer under that: limits you set yourself, aimed at exactly the traffic you want to control, so a single expensive model, a leaked key or one busy teammate cannot consume everything.

​
Rules

A rule is one set of targets, one quota type, one number, one reset behaviour. Create one from the Quotas page with the Create button, and change it later from the pencil on its row. The dialog describes the rule in plain language as you fill it in. A rule with No limit never refuses a request. Its usage is still counted and shown on the Quotas page, which is a way to watch a number before capping it.

​
Targets

A rule has four optional targets. Each one is either a specific value or any. Targets combine. A request is counted only when it matches every target you set, so a rule on the key ci-runner and the model claude-opus-5 counts that key’s calls to that model and nothing else. Leave every target on any to cap the whole workspace with one number. A model and a provider on the same rule have to agree: the model must belong to that provider. A rule on an API key counts only requests made with that key through the gateway. Chat and Computer do not use your API keys, so their usage is never counted against a key rule. Use a member rule to limit what one person uses everywhere.

​
Quota types

Each type pairs a metric with a window. Daily and monthly types reset, total types never do. The picker lists every type with its category and a short description, and can be searched or sorted by category or by metric. Tokens. Input, output, cached input, cached output, total, cached total. Daily, monthly and total. Requests, images, audio seconds. Daily, monthly and total. Spending. Daily, monthly and total, in US dollars. Searches. Daily searches and daily deep searches.

​
Rolling and Calendar

Calendar resets at a fixed boundary. A daily rule goes back to zero at 00:00, a monthly rule on the 1st. Rolling counts a moving window ending right now, 24 hours for a daily rule and 30 days for a monthly rule, so old usage falls off gradually instead of all at once. A total type is cumulative and never resets, so Reset does not apply and the dialog hides it.

​
Limits

​
Workspace budget

Separate from the rules above, each workspace carries a monthly spend cap. It defaults to EUR 10.00, and setting it to zero means no cap. The cap is enforced before the request runs, not after. The gateway estimates what the call will cost, reserves that amount against the workspace, and once the response is complete it settles the reservation against what was actually spent. A request that would push the workspace past its cap is refused rather than run and billed.
The reservation is an estimate, so a workspace close to its cap can see a request refused whose real cost would have fit. Raise the budget rather than retrying.
Spend is measured over the calendar month and reads from the metrics recorded for that workspace.

​
When a quota is hit

The request is refused with 429 and a message naming the rule and its reset time. Nothing is charged for a refused request. Raise the limit, disable the rule or wait for the window to reset.