Limits, Quotas, and Performance

These limits are intentionally explicit so you can size clients and plan retries.

Rate limits

Environment	Limit	Notes
Sandbox	100 requests/min per API key	Bursty tests may hit 429 quickly
Production	1,000 requests/min per API key	Contact support for higher quotas

When rate limited, the API returns 429 rate_limited.

Payload and request limits

• Max payload size: 256 KB per message
• Max poll max: 50
• Default request timeout (server-side): 30 seconds

Concurrency guidance

• Scale consumers horizontally.
• Prefer smaller polling batches with higher poll frequency to reduce “large batch + long processing” visibility issues.

Performance considerations

• Publish latency is typically dominated by network RTT.
• Poll latency may vary depending on queue depth.

---

Next: Getting Started · Reference · Changelog · Support