Plans & quotas
Every org is on a plan. The plan is the source of truth for what an org’s requests may do, and it clamps every request before it’s matched.
The catalog
Section titled “The catalog”| Basic | Pro | Enterprise | |
|---|---|---|---|
| Included requests / mo | 50,000 | 2,000,000 | Custom |
| Included bandwidth | 5 GB | 200 GB | Custom |
| Rate limit | 5 rps | 50 rps | 500 rps |
| Max concurrency | 10 | 200 | 2,000 |
| Overage | per-GB | per-GB | negotiated |
Plans are a fixed catalog (basic, pro, enterprise) defined in one place
and seeded into the database; the dashboard’s plan page is generated from it.
What a plan clamps
Section titled “What a plan clamps”- Allowed regions / IP types — a request targeting outside them is rejected
with
403. An org can be restricted to, say, EU residential only. - Max concurrency — the number of simultaneous in-flight requests. Exceeding
it returns
429. - Rate limit (rps) — sustained request rate.
- Default strategy — the routing strategy used when a request doesn’t specify one.
Per-org overrides
Section titled “Per-org overrides”Within plan limits, an org can carry overrides (e.g. a tighter default strategy or a narrower region allow-list) without changing plans. The effective config the data plane enforces is the plan merged with these overrides.
Concurrency & rate limiting
Section titled “Concurrency & rate limiting”In-flight requests are counted per org in Redis with atomic increment/decrement,
so the concurrency cap is enforced consistently across all core instances. When
you’re at the cap, new requests get 429 until in-flight work drains.