Rate limit reached during a coding session
Developers often cannot tell whether the limit is from the account, workspace, model route, or gateway.
Check account limits first, then lower concurrency and retry with a smaller context.A single runbook for the errors developers keep hitting in Claude Code: API key setup, Pro or Max versus API billing, Messages-compatible endpoints, rate limits, server errors, and overloaded routes.
Instead of copying community screenshots into the site, these cards turn the recurring Reddit and X patterns into stable, source-safe troubleshooting prompts.
Developers often cannot tell whether the limit is from the account, workspace, model route, or gateway.
Check account limits first, then lower concurrency and retry with a smaller context.A common confusion is assuming a Claude subscription automatically includes API credits.
Treat Claude subscription access and API billing as separate unless the provider page says otherwise.Long agent runs can hide whether the failure is provider overload, a gateway route issue, or one bad retry loop.
Log status code, request id, route, model, token size, and retry count before changing code.The fastest fix is to separate key problems, quota problems, server errors, and overload.
Your account, workspace, model, or route is sending more requests or tokens than allowed.
The provider or route returned a server-side failure. One retry may work; repeated failures need routing evidence.
The upstream service is under heavy load. This is usually not fixed by changing the API key.
Claude Code is not using the key you expect, or the key lacks access to the requested model or endpoint.
If a developer says Claude Code has an API key problem, check the active key, base URL, and billing source. Do not assume a subscription plan and API credits are the same account bucket.
Get API accessClaude Code-style clients usually need a Messages-compatible endpoint, a valid API key, and a model route that supports the requested workload.
Check docsThe request volume, token volume, or model route has crossed a limit. Check the active API key, workspace, model route, context size, and parallel agent runs before changing prompts.
Treat Claude subscription access and API billing as separate unless the official account or provider page says they are linked. Claude Code can use different authentication paths, so verify the active key and billing source.
Capture the request id, retry once with backoff, and check whether the same model route fails repeatedly. If it repeats, switch route or contact support with the exact time, model, and request id.
529 means the upstream service is overloaded. Wait, back off, reduce request size, or use a fallback route. Replacing the API key usually does not solve overload by itself.
Use official docs for final limits and error semantics. Use community threads as demand signals, not as the source of truth.