Plans, limits and usage
The organisation owns the plan, and every allowance is pooled across its workspaces. What each plan gives, which operation stops where, and why a per-minute limit is a different animal from a monthly one.
A workspace has no plan. The organisation does, and every allowance below is pooled across all of that organisation's workspaces. Five workspaces on one Pro plan share one 30 GB of storage, not five - which is the single most useful thing to know before you split content across workspaces.
What each plan allows
| Free | Pro | Enterprise | |
|---|---|---|---|
| Workspaces | 1 | 5 | unlimited |
| Members | 1 | 5 | unlimited |
| Pages | 5 | unlimited | unlimited |
| Storage | 1 GB | 30 GB | unlimited |
| Per file | 50 MB | 500 MB | 2047 MB |
| AI credits / month | 50,000 once | 5,000,000 | unlimited |
| API requests / month | 100,000 | 2,000,000 | unlimited |
| Bandwidth / month | 50 GB | 200 GB | unlimited |
| Remove branding | no | yes | yes |
| Cart and checkout | no | yes | yes |
| SSO | no | no | yes |
Free AI credits are a one-time grant, not a monthly allowance: the monthly reset deliberately skips free organisations, so the 50,000 do not come back next month. Paid plans refill.
An organisation can also carry its own limits that override the plan's - which is how a negotiated enterprise agreement is expressed without inventing a plan for it.
What happens when you reach one
Different limits stop different things, and the wording tells you which one you met.
- Pages - creating another page is refused: Page limit reached for the organization's plan.
- Storage - an upload that would cross the line is refused before any bytes move: Storage limit exceeded. Current: 812.4MB, Limit: 1024MB.
- AI credits - a call is refused when its estimate would cross, rather than halfway through generating.
- API requests and bandwidth - delivery answers 429: Monthly API request limit reached or Monthly bandwidth limit reached.
Storage and AI credits may be filled exactly. It is the operation that would cross the line that fails, not the one that lands on it. The others stop on reaching the number.
The one exception worth knowing
The monthly delivery block applies only to organisations without provisioned billing - free, with no subscription. An organisation that pays keeps serving past the number rather than having its site go dark mid-month. Usage is still counted, and still visible.
Rate limits are not plan limits
Monthly allowances are about how much you use in a billing period. Rate limits are about how fast, and they exist even on plans with unlimited everything.
The delivery route budgets the workspace first - 1,200 requests a minute - and the calling address second, at 600. Both are tunable per deployment.
The order is deliberate. A customer's site behind a CDN or a serverless region serves every visitor from a handful of addresses, so an address budget is really a budget for that customer's entire site, and a crawler warming a cold cache spends it in seconds. The workspace is what the traffic actually costs, so that is what is budgeted; the address cap stays as an abuse net, an order of magnitude wider.
Over the limit you get 429 with Retry-After, X-RateLimit-Remaining and X-RateLimit-Reset. Honour Retry-After rather than retrying immediately - a client that retries on a 429 without backing off is how one slow page becomes an outage.
Where to see your numbers
The organisation's usage page shows every pooled figure against its allowance, with a per-workspace breakdown underneath - so "who is using the storage" has an answer. It also shows media signings, which are counted but capped by no plan.
The same numbers come from one place in the code, which is why a gauge cannot disagree with what enforcement blocks on.
Next steps
- Workspaces - how workspaces sit inside an organisation.
- Media - the per-file upload cap, which is separate from the storage allowance.
- Delivery API - the route those rate limits apply to.