Guides
Rate Limits
Request limits, headers, and 429 backoff
Limits by plan
| Plan | Request Limit | Concurrency | Best For |
|---|---|---|---|
| Free | 10 requests/min | 2 concurrent | Testing & prototyping |
| Pro | 100 requests/min | 10 concurrent | Production apps |
| Enterprise | 1000 requests/min | 50 concurrent | Large-scale operations |
Response headers
Every response includes the rate-limit headers:
X-RateLimit-Limit: 100
X-RateLimit-Remaining: 87
X-RateLimit-Reset: 1714137600X-RateLimit-Reset is a Unix epoch timestamp.
Handling 429
When you hit 429 RATE_LIMIT_EXCEEDED, back off until the X-RateLimit-Reset epoch — don't retry blindly. Pair with exponential backoff for transient errors:
import httpx, time
def call(url: str, headers: dict, body: dict, max_attempts: int = 5):
for attempt in range(max_attempts):
r = httpx.post(url, headers=headers, json=body, timeout=60.0)
if r.status_code != 429:
r.raise_for_status()
return r.json()
reset = int(r.headers.get("X-RateLimit-Reset", "0"))
wait = max(reset - int(time.time()), 2 ** attempt)
time.sleep(wait)
raise RuntimeError("rate-limited after retries")For batch jobs, prefer fewer larger batches over many small ones — a 50-URL batch is one request against your rate limit, not fifty.