Home / Docs / Status and reliability

Status and reliability

How the gateway handles upstream failures, and what to do when something is degraded.

How failover works

Requests are distributed across upstream capacity. When one upstream returns an error or is overloaded, traffic moves to another rather than surfacing a hard failure — which is why a 529 here is rarer than calling a single provider directly.

What each failure means for you

You seeWhere the fault isWhat to do
529 overloadedUpstream capacityRetry with backoff; usually clears in seconds
500GatewayRetry once, then report with the request id
429Your account's concurrencyReduce parallelism — Rate limits
Slow first tokenUpstream queueing under loadExpected during peaks; streaming makes it visible sooner

Building for degradation

For an incident affecting your account, contact support@claudeapikey.dev with timestamps and a request id.