Skip to main content

Rate Limiting

Limits​

Stendly applies rate limits per API key to ensure fair usage. Limits vary by endpoint.

Exceeding Limits​

When a rate limit is exceeded, the API returns:

  • HTTP Status: 429 Too Many Requests
  • Header: Retry-After (seconds to wait)

Handling Rate Limits​

Automatic Retry​

All SDKs automatically retry on 429 with exponential backoff:

client = Client(api_key="...", max_retries=5)

Manual Handling​

from stendly import RateLimitError
import time

try:
    intent = client.intents.create(...)
except RateLimitError as e:
    wait = int(e.retry_after)
    print(f"Rate limited. Waiting {wait}s...")
    time.sleep(wait)
    # Retry

Best Practices​

  • Implement exponential backoff with jitter
  • Cache idempotent responses when possible
  • Batch requests instead of sending many at once
  • Monitor your usage in the dashboard