Rate limit

A rate limit is a cap on how many API requests a client can make per unit of time. AI Detector API's free tier is rate-limited to keep abuse manageable; paid tiers have higher limits and predictable behavior.

When you exceed a rate limit, AI Detector API returns HTTP 429 (Too Many Requests) with a Retry-After header telling you how long to wait. SDKs handle this automatically with exponential backoff, but if you're hitting the REST endpoint directly, your code should respect the header.

Plan-by-plan limits. Free tier is 1,000 requests per calendar month with a soft per-second cap. Pro is 50,000 requests per month with a 50 requests per second burst. Enterprise is custom, typically uncapped for typical workloads with a separate burst negotiation. The monthly cap is reset on the first day of your billing cycle at 00:00 UTC.

Response headers. Every API response includes three rate-limit headers you can read programmatically:

X-RateLimit-Limit: 50000
X-RateLimit-Remaining: 12389
X-RateLimit-Reset: 1735689600

Read X-RateLimit-Remaining periodically to surface usage to your team before you hit the cap. The Reset value is a Unix timestamp.

Handling 429s in code. The official SDKs retry with exponential backoff and jitter automatically up to three times. If you are hitting the REST endpoint directly, respect Retry-After:

import time, requests

while True:
    r = requests.post(url, headers=headers, json={"text": text})
    if r.status_code != 429:
        break
    retry_after = int(r.headers.get("Retry-After", "1"))
    time.sleep(retry_after)

When you legitimately need more. If you are hitting limits because of bursty traffic (start of business day, mass batch upload), the cleanest fix is the batch endpoint (one request per 100 documents) plus webhooks. If your steady-state volume just exceeds the Pro tier, contact us for Enterprise pricing.