> ## Documentation Index
> Fetch the complete documentation index at: https://unkey.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Unkey is two separate products. Compute builds, deploys, and runs apps behind a gateway. API Management issues API keys, enforces rate limits, manages identities and permissions, and reports usage. Say which product a page belongs to; a reader can use either without the other.
> Every Unkey API endpoint is an HTTP POST to https://api.unkey.com/v2/{service}.{procedure} with a root key in the Authorization: Bearer header. Root keys are workspace scoped.
> Error codes have the form err:{system}:{category}:{specific} and each has a page at /errors/{system}/{category}/{specific}.
> The word environment means production or preview in Compute. Rate limiting has four meanings on this site; the glossary lists them.

# ratelimits tables

> Every column in the rate limit tables and their rollups.

Query <Tooltip tip="Here: decisions made by the standalone ratelimit API (ratelimit.limit and ratelimit.multiLimit). Not key limits checked during verification and not the gateway policy.">rate limit</Tooltip> decisions with `analytics.getRatelimits` from five tables. They record calls to `ratelimit.limit` and `ratelimit.multiLimit`. Limits on keys and identities show up as `RATE_LIMITED` outcomes in the [key\_verifications tables](/docs/api-management/analytics/tables/key-verifications) instead.

## ratelimits\_v1

One row per rate limit check.

<ResponseField name="request_id" type="String">
  The `meta.requestId` of the request that performed the check.
</ResponseField>

<ResponseField name="time" type="Int64">
  When the check happened, in milliseconds since the Unix epoch.
</ResponseField>

<ResponseField name="inserted_at" type="Int64">
  When the row was recorded, in milliseconds since the Unix epoch. Filter on `time` instead.
</ResponseField>

<ResponseField name="workspace_id" type="String">
  Your workspace. Always the same value.
</ResponseField>

<ResponseField name="namespace_id" type="String">
  The namespace the check ran in, as an `rlns_` ID.
</ResponseField>

<ResponseField name="identifier" type="String">
  The identifier you passed: a user ID, an IP, a tenant, whatever you chose to limit on.
</ResponseField>

<ResponseField name="passed" type="Bool">
  `true` if the request was allowed, `false` if it was rejected.
</ResponseField>

<ResponseField name="latency" type="Float64">
  How long Unkey took to process the check, in milliseconds.
</ResponseField>

<ResponseField name="override_id" type="String">
  The override that applied, or empty when the request used the default limit.
</ResponseField>

<ResponseField name="limit" type="UInt64">
  The limit in force for this check.
</ResponseField>

<ResponseField name="remaining" type="UInt64">
  Requests left in the window after this check.
</ResponseField>

<ResponseField name="reset_at" type="Int64">
  When the window resets, in milliseconds since the Unix epoch.
</ResponseField>

<ResponseField name="tokens" type="UInt64">
  The `cost` charged, `1` by default. Zero when the request sent `cost: 0`, or for older rows recorded before this column existed.
</ResponseField>

## Rollup tables

`ratelimits_per_minute_v1`, `ratelimits_per_hour_v1`, `ratelimits_per_day_v1`, and `ratelimits_per_month_v1` have the same columns: one row per time bucket, namespace, and identifier.

<ResponseField name="time" type="DateTime | Date">
  Start of the bucket. `DateTime` on per-minute and per-hour, `Date` on per-day and per-month.
</ResponseField>

<ResponseField name="workspace_id" type="String">
  Your workspace.
</ResponseField>

<ResponseField name="namespace_id" type="String">
  The namespace `rlns_` ID.
</ResponseField>

<ResponseField name="identifier" type="String">
  The rate limited identifier.
</ResponseField>

<ResponseField name="passed" type="SimpleAggregateFunction(sum, Int64)">
  Checks that were allowed in the bucket. Read with `sum(passed)`.
</ResponseField>

<ResponseField name="total" type="SimpleAggregateFunction(sum, Int64)">
  All checks in the bucket. `sum(total) - sum(passed)` is the number rejected.
</ResponseField>

<ResponseField name="total_tokens" type="SimpleAggregateFunction(sum, Int64)">
  Sum of `tokens` over all checks.
</ResponseField>

<ResponseField name="passed_tokens" type="SimpleAggregateFunction(sum, Int64)">
  Sum of `tokens` over allowed checks.
</ResponseField>

<ResponseField name="latency_avg" type="AggregateFunction(avg, Float64)">
  Can't be read, because `avgMerge` isn't allowed. Use `avg(latency)` on the raw table.
</ResponseField>

<ResponseField name="latency_p75" type="AggregateFunction(quantilesTDigest(0.75), Float64)">
  Aggregate state, with the same limitation.
</ResponseField>

<ResponseField name="latency_p99" type="AggregateFunction(quantilesTDigest(0.99), Float64)">
  Aggregate state, with the same limitation.
</ResponseField>

The rate limit tables keep less history than the verification tables: raw 1 month, per-minute 7 days, per-hour 30 days, per-day 100 days, per-month 3 years. For a month-long chart, use the per-day table. Your plan's retention also applies. See [Analytics restrictions and quotas](/docs/api-management/analytics/restrictions-and-quotas).