Rate Limits

Rate limits are enforced per API key and vary by tier. All limits use a sliding window. Exceeding limits returns a 429 status with a Retry-After header.

Limits by Tier

Developer

Free
Requests / Minute60 RPM
Requests / Day1,000
Max Context Length8,192 tokens
Concurrent Connections2
Providers AvailableLlama-3, DeepSeek
FailoverSingle fallback

Professional

Pro
Requests / Minute1,000 RPM
Requests / Day50,000
Max Context Length32,768 tokens
Concurrent Connections10
Providers AvailableAll (GPT-4o, Claude, etc.)
Failover2-deep cascade

Enterprise

Enterprise
Requests / Minute10,000+ RPM
Requests / DayUnlimited
Max Context Length200,000 tokens
Concurrent ConnectionsUnlimited
Providers AvailableAll + Priority Queue
Failover3-deep cascade + Hot Standby

Response Headers

Every API response includes rate limit headers so you can monitor usage programmatically.

X-RateLimit-Limit

Maximum number of requests allowed in the current window.

X-RateLimit-Remaining

Number of requests remaining before hitting the limit.

X-RateLimit-Reset

Unix timestamp (seconds) when the rate limit window resets.

Retry-After

Seconds to wait before retrying (only present on 429 responses).

Best Practices

  • •Implement exponential backoff when receiving 429 responses.
  • •Monitor X-RateLimit-Remaining to proactively throttle before hitting limits.
  • •Use batch endpoints for bulk workloads instead of sending individual requests.
  • •Contact the Xypherar team for custom Enterprise limits tailored to your throughput needs.
Need Higher Limits?

Join the Genesis Pilot Queue

Get priority access to Professional and Enterprise tiers with custom rate limits.

Error CodesBack to Docs