FAQ: What should I do about rate limits?

NexoRouter can return rate limit errors when request volume or estimated token volume exceeds the current gateway window.

Error codes

Code	Meaning
`rate_limit_exceeded`	Too many requests in the current RPM window.
`token_rate_limit_exceeded`	Too many estimated tokens in the current TPM window.

Both errors can include retry-after and rate-limit headers.

Fix it

Wait for the retry window.
Reduce concurrency.
Batch less aggressively.
Shorten input.
Contact support before sustained production load.

Do not retry immediately in a tight loop. Rate-limited requests are rejected before they reach the provider, so immediate retries usually just repeat the 429 until the window resets.