Rate limits and quotas
| Plan | APIs | Calls | Rate limit per key | Keys | Memory facts | Spatial spaces |
|---|
| Free | Verification | 1,000 per month | 2/s, burst 10 | 1 | - | - |
| Trial | Verification, Memory, Signals, Spatial Memory | 25,000 in 14 days | 10/s, burst 40 | 2 | 100 | 2 (≤ 20,000 points) |
| Developer | Verification, Memory, Signals, Spatial Memory | 25,000 per month | 10/s, burst 40 | 2 | 250 | 5 (≤ 50,000 points) |
| Team | Verification, Memory, Signals, Spatial Memory | 250,000 per month | 50/s, burst 200 | 10 | 1,000 | 25 (≤ 200,000 points) |
Response headers
X-RateLimit-Limit | Requests per second allowed for this key. |
X-RateLimit-Burst | Largest burst allowed. |
X-RateLimit-Remaining | Requests left in the current burst. |
X-Quota-Limit | Calls allowed in the current window. |
X-Quota-Remaining | Calls left in the current window after this one. |
X-Quota-Reset | When the window renews (UTC). |
Retry-After | On 429 rate_limited: seconds to wait. |
What counts
A call is one authenticated request to the Verification, Memory, Spatial Memory or Signals API that we answered. Requests refused with 429, requests that failed with a 5xx status, and requests the gateway refuses before they reach an API (for example an API your plan does not include, or a body over 10 MB) are not counted. /v1/account and /v1/usage are not counted but are rate-limited.