Rate limits - Neural Draft

API keys

Every /v1 request with an API key counts against 60 requests per minute per key. Reads and writes count the same. A long-running job (blog post, translation) counts once when you submit it, and an SSE stream on /v1/jobs/{id}/stream counts as one request. Every response carries:

Header	Meaning
`X-RateLimit-Limit`	Requests allowed per window (60).
`X-RateLimit-Remaining`	Requests left.
`X-RateLimit-Reset`	Unix time when the window resets.

Over the limit you get 429:

{
  "type": "https://neuraldraft.com/errors/rate-limit-exceeded",
  "title": "Too Many Requests",
  "status": 429,
  "detail": "Rate limit exceeded. Retry after 14 seconds.",
  "code": "rate_limit_exceeded",
  "retry_after": 14
}

Wait retry_after seconds, then retry (see Errors → Retry strategy). Need more for a known workload? Email info@neuraldraft.io.

Public endpoints (no key)

Limited per visitor IP:

Endpoint	Limit
`/v1/public/availability`, `/v1/public/services/{id}`, `/v1/public/bookings`, `POST /v1/contact-forms`, `POST /v1/newsletters/subscribe`	5 / minute
`POST /v1/checkout-sessions`	5 / minute
`/v1/public/products`, `/v1/public/product-categories`	60 / minute
`/v1/public/orders/{number}`	30 / minute
`/v1/public/calendar/{token}.ics`	60 / minute

nd.js blocks on a site (/_nd/api/*) have their own limits per IP and workspace: forms 5 a minute and 30 an hour, newsletter sign-ups 5 a minute, bookings 5 a minute and 20 an hour.