Rate limits

Per-plan tiers

Limits are tied to the subscription on the account that owns the API key. See pricing for plan details.

Response headers

Every API response includes:

X-RateLimit-Limit: 60
X-RateLimit-Remaining: 47
X-RateLimit-Reset: 1715797260

• Limit — Your current per-minute ceiling.
• Remaining — Requests left in the current 60-second window.
• Reset — Unix timestamp (seconds) when the window resets and Remaining returns to Limit.

When you hit 429

Exceeding the limit returns:

HTTP/1.1 429 Too Many Requests
Retry-After: 23
X-RateLimit-Limit: 60
X-RateLimit-Remaining: 0
X-RateLimit-Reset: 1715797260

{
  "success": false,
  "error": {
    "code": "AUTH_RATE_LIMITED",
    "message": "Rate limit exceeded. Try again in 23 seconds.",
    "details": { "retry_after_seconds": 23 }
  },
  "meta": { "request_id": "req_..." }
}

Respect Retry-After (in seconds). Aggressive retry before the window resets will keep you locked out and counts as part of the per-day abuse signal.

Client-side rate-limit handling

TypeScript

async function callWithBackoff(url: string, body: unknown) {
  while (true) {
    const res = await fetch(url, {
      method: "POST",
      headers: {
        Authorization: `Bearer ${process.env.REELSBUILDER_API_KEY}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify(body),
    });
    if (res.status !== 429) return res;
    const retryAfter = Number(res.headers.get("retry-after") ?? 1);
    await new Promise((r) => setTimeout(r, retryAfter * 1000));
  }
}

Python

import os, time, requests

def call_with_backoff(url, body):
    while True:
        r = requests.post(
            url,
            headers={"Authorization": f"Bearer {os.environ['REELSBUILDER_API_KEY']}"},
            json=body,
            timeout=30,
        )
        if r.status_code != 429:
            return r
        time.sleep(int(r.headers.get("retry-after", "1")))

Concurrent jobs

Beyond per-minute request limits, async generation jobs have a concurrency cap (see table above). If you queue more than your concurrent-job ceiling, extra jobs return 202 Accepted with a queue_position field — they start as in-flight jobs complete.

Burst handling

The limiter uses a sliding 60-second window, not a fixed clock minute. Spreading 60 requests over 60 seconds is safer than firing all 60 at the start of a clock minute. If you must burst, pause for X-RateLimit-Reset - now() seconds after every full window.

Need a higher limit?

Most workloads fit comfortably inside the Business tier. For sustained higher throughput, custom routing, or dedicated capacity: contact admin@reelsbuilder.ai with your projected RPM and use case.

See also

• Errors — full reference for AUTH_RATE_LIMITED and related codes
• Idempotency — safe retry on 429 with backoff