Rate limits explained
Why the API answers “too many requests” when a key sends too many in a minute, and what to do about it.
Every API key can make a limited number of requests per minute. The limit protects your account: one script that goes wrong cannot slow Filmo down for your team or for the other tools that use the API.
What the limits are
| What | Limit | Applies to |
|---|---|---|
| Any request | 120 per minute | Each API key, across everything it does |
| Availability checks | 30 per minute | Each API key, for items, crew and vehicles together |
Availability checks have their own, smaller limit because they are the heaviest questions Filmo answers, and they also count in the general limit. The limits are the same on every plan.
An AI assistant connected to Filmo is subject to the same limits: each thing it looks up or changes is a request on the connection it works with.
What happens when a key goes over
Filmo does not break anything and does not lose data. It answers the extra requests with “too many requests” and the number of seconds to wait. When the minute is over, the key works again.
If an integration or an assistant says “too many requests”
- Wait a minute. The limit counts per minute, so most of the time that is all it takes.
- Check what is using the key. Open the API and MCP page of Settings, then the tab for API keys, and look at the key's name and when it was last used. A key shared by several tools adds all their requests together.
- Give each tool its own key. Then one busy tool cannot use up another one's requests. See Create an API key.
- Ask the person who built the integration to spread out its requests. A well-built integration waits the time Filmo asks for and asks only for what changed, instead of reading everything again and again.
Bringing in a lot of data at once
A one-off import of thousands of records can reach the limit on purpose. Spread it over time, or ask whoever is doing it to follow the developer guide below.
For the person who builds the integration
Every response tells the integration how many requests it has left, and a refused one says how long to wait. The details, with retry examples, are in the rate limits guide, and the limits that matter for AI assistants in the MCP limits guide.
- Choose what a key may do: API permissions (scopes).
- Keep another system up to date: React to changes without webhooks.
Frequently asked questions
The limits are the same for every account. If your integration needs more, contact support and tell us what it does.
No. They apply to API keys, and to the AI assistants connected through them, not to people working in the app.