Limits and usage
Every team has a plan. The plan sets how many checks the team can make each day and each month, how fast it can send requests, how long it keeps history, and whether it can buy extra checks. Plans can change their limits at any time, and a change reaches your team within a minute.
| Plan | Checks a month | Checks a day | Requests per second | Burst | History | Extra checks |
|---|---|---|---|---|---|---|
| Free | 2,000 | 100 | 2 | 10 | 3 days | No |
| Pro | 20,000 | No limit | 10 | 50 | 30 days | $1.50 per 1,000 |
| Scale | 200,000 | No limit | 50 | 250 | 90 days | $0.70 per 1,000 |
| Enterprise | Agreed | Agreed | 100 | 500 | 365 days | Agreed |
What counts
- Each text check counts as one check, including checks answered from your rules or from the cache.
- Each item of a batch counts as one check.
- Checks made with a test key (
sg_test_) never count. - Checks that fail, for example with
503 model_unavailable, never count.
Quotas
The daily quota resets at 00:00 UTC. The monthly quota resets when your billing period starts again, or on the first of each month (UTC) on Free.
When a check would go over a quota, it fails with 429 and a retry-after header that says how many seconds until the quota resets:
name |
Status | Meaning |
|---|---|---|
rate_limit_exceeded |
429 | Too many requests at once. Wait the seconds in retry-after. |
daily_quota_exceeded |
429 | The team used all of today's checks. They reset at 00:00 UTC. |
monthly_quota_exceeded |
429 | The team used all of this period's checks, and pay-as-you-go is off. |
overage_cap_reached |
429 | Pay-as-you-go checks reached the team's cap for this period. |
capacity_paused |
503 | Checks that need the AI model are paused briefly. Wait the seconds in retry-after. |
Rate limits
Each team has one rate limit, shared by all of its keys. You can send up to the burst size at once, and the allowance refills at the plan's requests per second. Every response to a request with an API key carries these headers:
| Header | Meaning |
|---|---|
ratelimit-limit |
The most requests you can send at once (the burst). |
ratelimit-remaining |
How many you can send right now. |
ratelimit-reset |
Seconds until the full burst is available again. |
ratelimit-policy |
The long-run rate, for example 10;w=5: 10 requests per 5 seconds. |
retry-after |
Only on 429: seconds to wait before trying again. |
A batch request counts as one request against the rate limit, and each of its items counts as one check against your quotas.
Pay-as-you-go
On Pro and Scale, team admins can turn on pay-as-you-go in the dashboard. With it on, checks beyond the monthly quota keep working and are billed per 1,000 checks. They stop at a cap, 4 times the monthly quota by default, which admins can set anywhere from 1 to 10 times. At the cap, checks fail with 429 overage_cap_reached until the period resets or an admin raises the cap.
Usage events
Your team gets one event per period for each of these, as a webhook and an email to admins:
usage.threshold_reachedat 80% of the daily or monthly quotausage.quota_exceededat 100% of the daily or monthly quotausage.overage_cap_reachedwhen pay-as-you-go reaches its cap
Read your usage
GET /v1/usage
Authorization: Bearer sg_live_...This needs a full-access key. The numbers are the same counters that enforce your quotas, and the same ones the dashboard's Usage page shows.
{
"object": "usage",
"plan": "pro",
"period": {
"start": "2026-11-01T00:00:00.000Z",
"end": "2026-12-01T00:00:00.000Z"
},
"monthly": {
"used": 20003,
"limit": 20000,
"resets_at": "2026-12-01T00:00:00.000Z"
},
"daily": {
"used": 45,
"limit": null,
"resets_at": "2026-11-15T00:00:00.000Z"
},
"rate_limit": { "requests_per_second": 10, "burst": 50 },
"overage": {
"available": true,
"enabled": true,
"used": 3,
"cap": 80000,
"cap_multiplier": 4,
"price_per_1000_usd": 1.5
}
}| Field | Meaning |
|---|---|
plan |
The team's plan. |
period |
The current monthly quota period. |
monthly.used, daily.used |
Checks counted so far in the period and today. |
monthly.limit, daily.limit |
The quotas. null means no limit of that kind. |
resets_at |
When each quota starts again. |
rate_limit |
The team's requests per second and burst. |
overage.available |
Whether the plan offers pay-as-you-go. |
overage.enabled |
Whether an admin turned it on. |
overage.used |
Checks beyond the monthly quota this period. |
overage.cap |
Where pay-as-you-go stops, in checks. null when it is off. |
overage.price_per_1000_usd |
The price of 1,000 extra checks. |
Free plan
To keep the free plan useful for everyone:
- You need a verified email address to create a team or an API key.
- Each person can have at most 2 teams on the Free plan.
- Free keys that send many near-identical messages that score as spam are flagged for review.