Quota Management

Quota Management controls spending through the UURoute API. It helps individual developers and teams manage costs, isolate usage across applications, and reduce the risk associated with leaked API keys or abnormal requests.

UURoute supports two quota levels:

Quota level	Applies to	Primary purpose
API Key quota	An individual Inference Key	Set an independent spending limit for an application, environment, or service
Org member quota	An individual member of an organization	Let an Org admin control each member’s monthly spending at the Credit level

This guide explains how to configure API Key quotas and Org member quotas, what each quota applies to, and how UURoute evaluates them when both are enabled. For information about creating organizations, managing members and permissions, and using Org Credits, see the Organization documentation.

How API Key Quotas Work

An API Key quota appears as Quota Limit in the UURoute console. When enabled, UURoute tracks the cost generated by that key during the current quota period and compares it with the configured Quota Limit.

A quota applies only to the API key on which it is configured. It does not affect other keys under the same account. For example, you can create separate keys for development, testing, and production and assign a different quota to each one.

The Quota column in the API key list displays the amount used in the current period, the quota limit, and the refresh period. For example:

$0 / $100  Daily refresh

This means the key has a $100 quota that refreshes daily and has used $0 during the current period.

A quota limits spending. It does not add, reserve, or increase Credits. Your account or organization must still have enough available Credits to make API requests.

Create an API Key with a Quota

You can configure a quota when creating an Inference Key:

Go to Workspace → Inference Keys.
Click Create.
Enter a name for the key.
Set an Expiration Time if needed. Leave it blank if the key should not expire automatically.
Enable Quota Limit.
Enter the limit under Quota Limit ($).
Select a refresh period under Quota Refresh.
Save the key and store the newly created API key securely.

The Quota Limit must be a whole number greater than 0 and less than 9,990,000. The currency is US dollars (USD).

Quota Refresh Periods

Quota Refresh determines when usage accounting starts over. All refresh periods use Coordinated Universal Time (UTC) rather than the local time zone of the user or browser.

Option	Refresh time in UTC	Recommended use case
Never	Does not refresh automatically; costs continue to accumulate	One-time tasks, proofs of concept, or keys that need a lifetime spending cap
Daily	00:00 UTC every day	Daily cost controls for high-frequency services
Weekly	00:00 UTC every Monday	Development teams or recurring batch jobs
Monthly	00:00 UTC on the first day of each month	Monthly cost controls for production applications, departments, or customers

A refresh resets only the amount used during the current quota period. It does not add funds to the account or extend the key’s Expiration Time.

The Quota information in the API key list displays the next refresh time, for example Next reset at 2026-08-20 00:00:00 UTC. Use the displayed UTC time as the source of truth. When converting it to local time, account for the local time zone and daylight saving time. For example, 00:00 UTC corresponds to 08:00 on the same day in China Standard Time (UTC+8).

View an API Key Quota

Go to Workspace → Inference Keys. The key list displays:

Quota: Cost used during the current period, quota limit, and refresh period.
Creation Time: When the key was created.
Expiration Time: When the key expires.
Status: Whether the key is currently enabled.

For a more detailed cost analysis, go to:

Workspace → Usage: Review requests, tokens, models, providers, and costs by API key.
Workspace → Logs: Review request error logs for the current user.
Workspace → Dashboard: Review spending and request trends for the current user.

Usage data can be subject to a short delay. A recently completed request may not appear immediately on the Quota or Usage page.

Modify an API Key Quota

To update the quota for an existing key:

Go to Workspace → Inference Keys.
Find the key and open the ··· action menu on the right.
Open the edit page.
Enable or adjust Quota Limit, then set the new amount and refresh period.
Save your changes.

Before lowering a quota, check the amount already used in the current period. If the new limit is lower than the amount already used, the key may be unable to make further requests because its quota has been exhausted.

Disable an API Key Quota

If a key no longer needs an individual spending limit, disable Quota Limit on its edit page and save the change. The key will no longer be restricted by its own quota, but it remains subject to:

Available account or Org Credits.
Org member quota.
The key’s status and Expiration Time.
Any other applicable platform usage or security limits.

Disabling the API Key quota does not make API usage unlimited or free.

API Key Quotas, Org Member Quotas, and Credits

These settings control different aspects of API access:

Setting	What it controls	Effect when exhausted or reached
API Key quota	The amount an individual key can spend during a quota period	The key may be unable to make further requests until the quota refreshes or its limit is changed
Org member quota	The amount an organization member can spend each month, measured in Credits	When the member’s quota is exhausted, that member’s keys in the organization may be unable to make further requests
Credits	The actual available balance of a personal account or organization	Requests cannot be billed normally when the relevant balance has insufficient Credits
Expiration Time	How long a key remains valid	The key becomes invalid when it reaches its expiration time

A single request may be subject to several limits at the same time. Even if an API Key quota has remaining capacity, the request may still fail if there are insufficient Credits, the Org member quota is exhausted, or the key is disabled or expired.

Org Member Quotas

Within an organization, an Org admin can configure a member’s Org member quota under Organization → Members. This quota is measured in Credits, and its console field is named credit monthly limit. Members without an individually configured limit use the default Org member quota defined under Organization → Settings.

An API Key quota and an Org member quota can apply at the same time:

An API Key quota limits only that key.
An Org member quota limits the combined monthly Credit usage of all keys belonging to that member within the organization.
Org Credits determine whether the organization has a sufficient balance to pay for requests.

For example, suppose a member has an Org member quota of 500 Credits and one of their API keys has a monthly quota of $100. That key is subject to both its own $100 quota and the member quota. All keys belonging to the member count toward the shared 500-Credit Org member quota. Exhausting any applicable limit may prevent the relevant requests from continuing.

For complete information about organization funding, member management, permissions, and quota allocation, see the Organization documentation.

Best Practices

Use separate keys for each application and environment: Create different keys for development, testing, and production so their usage is not mixed together.
Start with a smaller quota: Increase the quota gradually based on actual spending trends in Usage to reduce the risk of unexpected charges.
Use Never for experiments: A non-refreshing lifetime limit is useful for one-time tests or batch jobs.
Use periodic quotas for production: Choose Daily, Weekly, or Monthly based on your application’s usage pattern to prevent a single incident from consuming funds indefinitely.
Set an expiration time as well: Configure an Expiration Time for temporary projects, demos, and external collaborations.
Review Usage and Logs regularly: Quotas limit costs but do not replace usage analysis, troubleshooting, or secure key management.
Revoke leaked keys immediately: Do not rely on a quota alone to limit the impact of a leaked key.

FAQ

What happens if Quota Limit is not enabled?
The key does not have an individual quota, but it is still subject to Credits, Org member quota, key status, Expiration Time, and other applicable limits.

Does refreshing a quota add Credits?
No. A refresh starts a new quota accounting period but does not change the actual balance of the account or organization.

Which time zone is used for quota refreshes?
All quota refresh periods use UTC. Daily quotas refresh at 00:00 UTC each day, Weekly quotas at 00:00 UTC each Monday, and Monthly quotas at 00:00 UTC on the first day of each month. The next refresh time appears in the API key’s Quota information.

Why does an API request fail when the quota still has capacity remaining?
Check the available Personal account or Org Credits, Org member quota, key status, Expiration Time, and request error logs. Any applicable limit can cause a request to fail.

Does modifying a quota create a new API key?
No. Changing the Quota Limit or refresh period does not change the key itself.

Can a member bypass their Org member quota by creating multiple keys?
No. Each key can have its own quota, but all keys belonging to the member within the organization remain subject to the shared Org member quota.