OpenAI API 429 Rate limit reached for requests: what it means and how to fix it

What it means

You have hit your assigned rate limit: too many tokens or requests were submitted in a short period. Limits are measured in RPM, RPD, TPM, TPD, IPM and apply at organization and project level, not per user. The Python library raises RateLimitError.

Causes

• A loop or script that makes frequent or concurrent requests
• Sharing your API key with other users or applications
• Using a free plan that has a low rate limit
• Reaching the defined limit on your project

How to fix it

1. 1
   Fix 1.

   Pace your requests and avoid unnecessary or redundant calls.

2. 2
   Fix 2.

   If a Retry-After header is present, wait at least that long before retrying; if missing, use exponential backoff with jitter and limit the number of retries. Official SDKs automatically retry eligible 429 responses.

3. 3
   Fix 3.

   Set max_tokens as close to your expected response size as possible, since the rate limit counts the maximum of max_tokens and the estimated request tokens.

4. 4
   Fix 4.

   Use the Batch API for work that does not need immediate responses, upgrade usage tier, or ask your organization owner to raise project rate limits.

5. 5
   Fix 5.

   Configure SDK retries (default is 2).

   Snippet

   Copy

   client = OpenAI(
       # default is 2
       max_retries=0,
   )