llm rate_limit_error ai_generated true

Error 429: Too Many Requests - Rate limit exceeded

ID: llm/rate-limit-429

Also available as: JSON · Markdown
88%Fix Rate
90%Confidence
3Evidence
2023-01-01First Seen

Version Compatibility

VersionStatusIntroducedDeprecatedNotes
any active — — —

Root Cause

LLM API rate limit exceeded (requests/min or tokens/min).

generic

Workarounds

  1. 92% success Implement exponential backoff with jitter 
    wait = min(base * 2**attempt + random.uniform(0,1), max_wait)
  2. 85% success Use token bucket or leaky bucket rate limiter 
    Proactively throttle requests to stay under limit

Dead Ends

Common approaches that don't work:

  1. Retry immediately without backoff 90% fail

    Hammering the API worsens rate limiting and may cause IP blocks.

  2. Switch API keys rapidly to bypass limits 75% fail

    Most providers rate-limit per organization, not per key.