Docs

Rate limits

Per-second and per-month quotas. Headers on every response let you back off gracefully.

Per-category rate limits (req/sec)

Limits are enforced per API key and endpoint category, so separate keys do not contend with each other. The backend allows a short burst up to 2x the configured per-second rate and then refills at the plan rate.

Developer : 5/sec lightweight / 2/sec standard / 1/sec medium / 1/sec heavy / — AI

Starter : 20/sec lightweight / 10/sec standard / 5/sec medium / 5/sec heavy / 2/sec AI

Growth : 100/sec lightweight / 50/sec standard / 25/sec medium / 20/sec heavy / 10/sec AI

Business Intelligence: 400/sec lightweight / 200/sec standard / 120/sec medium / 100/sec heavy / 50/sec AI

Enterprise : Dedicated negotiated capacity

Weighted API units (monthly quota)

Not every call costs the same. Units are stored as integer tenths internally to avoid float drift.

Lightweight (codes, currencies, UQC) — 0.1 unit

Standard (HSN, ports, EDI, events) — 1 unit

Medium (scheme codes, RoDTEP) — 3 units

Heavy (sanctions screening) — 5 units

AI (impact-analyze, semantic search) — 20+ units

Response headers

Every 2xx and 429 response carries:

X-RateLimit-Category: standard
X-RateLimit-Limit: 30
X-RateLimit-Remaining: 27
X-RateLimit-Reset: 1748376000
X-Quota-Period-Used: 187231
X-Quota-Period-Limit: 500000
X-Unit-Weight: 1
Retry-After: 2

Backoff

On 429, retry after the Retry-After seconds. Exponential backoff with jitter is recommended.