> ## Documentation Index
> Fetch the complete documentation index at: https://docs.originalis.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Rate Limits & Budgets

> The request limit, the actions budget, and the headers that report both.

## Request limit

Each key has a daily request budget (default 5,000/day, resets at UTC
midnight). Successful responses report where you stand:

| Header                  | Meaning                                        |
| ----------------------- | ---------------------------------------------- |
| `X-RateLimit-Limit`     | The key's daily budget                         |
| `X-RateLimit-Remaining` | Requests left today                            |
| `X-RateLimit-Reset`     | Seconds until the budget resets (UTC midnight) |

`429` responses carry `Retry-After` (seconds). The headers are omitted on
the rare request where the counter is unreachable — they are never
guessed.

## Actions budget

Actions run real analysis pipelines with real cost, so each key also gets
a separate actions budget (default 25/day). It is charged **only when a
run actually dispatches** — rejected, deduplicated, and conflicting
requests cost nothing. Over-budget returns `429` with `Retry-After` plus
`X-Actions-Limit` / `X-Actions-Remaining`.

Contact us if your integration needs higher caps.