Rate limit 429 errors spike after migrating to gpt-4o-mini

Marcus WebbSep 12, 2024

After migrating from gpt-3.5-turbo to gpt-4o-mini, I'm seeing a massive spike in 429 rate limit errors even though my request volume hasn't changed.

My setup:

Tier 2 account

~200 RPM average

Using the Python SDK with retry logic

Error I'm getting:

openai.RateLimitError: Error code: 429 - Rate limit reached for gpt-4o-mini

I assumed gpt-4o-mini would have the same or higher rate limits as gpt-3.5-turbo. Is this not the case? Has anyone else experienced this?

4.3k views23 replies45 likesSolved

3 Replies

Karen WuStaffAccepted AnswerSep 13

Thanks for reporting this. GPT-4o-mini has separate rate limits from GPT-3.5 Turbo. You may be on a lower tier for this specific model. Check your dashboard at platform.openai.com for model-specific limits.

Marcus Webb Sep 14

That was it! I was on Tier 2 for GPT-3.5 but only Tier 1 for GPT-4o-mini. After spending $50 more I got bumped up. Thanks @karen_oai!

Nina Patel Sep 13

Same issue here. The tier system per model is confusing. Would be great if the error message mentioned the specific model's tier.