How do rate limits work?
API is rate-limited per-key: 60 requests/minute on Pro, 300/min on Business, 1,000/min on Agency. 429 response when exceeded.
Last updated: 2026-05-04
Per-tier limits
| Tier | Requests/minute | Burst (10s) |
|---|---|---|
| Pro | 60 | 100 |
| Business | 300 | 500 |
| Agency | 1,000 | 1,500 |
Exceeding the rate limit returns:
HTTP/1.1 429 Too Many Requests
Retry-After: 12 X-RateLimit-Limit: 60 X-RateLimit-Remaining: 0 X-RateLimit-Reset: 1730743212</code></pre>
Burst handling
We use a token-bucket algorithm. If you make 100 requests in 10 seconds, you're fine — even on Pro tier. But if you sustain >60/min, the next request gets rate-limited.
Best-practice client behavior
for request in batch:
response = call_aict_api(request) if response.status_code == 429: retry_after = int(response.headers['Retry-After']) time.sleep(retry_after) response = call_aict_api(request) # retry once yield response.json()</code></pre>
Most language SDKs handle this automatically.
Concurrency limits
In parallel, max simultaneous requests:
- Pro: 5
- Business: 25
- Agency: 100
Attempts beyond these queue (up to 30s, then 503). To get more, contact sales.
Why rate limits?
Protects shared backend infrastructure. Without limits, one runaway script could starve other paying users.
Was this helpful?
Still stuck? Contact support
