OpenAI API error 429 rate limit reached

You're building with the OpenAI API and suddenly hit a wall: Error 429: Rate limit reached for requests. Your requests stop going through, your app stalls, and you're left wondering what just happened and how to fix it.

OpenAI API error 429 rate limit reached

You're building with the OpenAI API and suddenly hit a wall: `Error 429: Rate limit reached for requests`. Your requests stop going through, your app stalls, and you're left wondering what just happened and how to fix it.

What's actually happening

Error 429 means you've exceeded OpenAI's rate limits — the maximum number of requests or tokens you can send in a given time window. OpenAI sets these limits per model and tier to prevent abuse and ensure fair access for everyone.

The limits work on multiple levels. You've got requests per minute (RPM), tokens per minute (TPM), and sometimes requests per day (RPD). When you cross any threshold, the API refuses your request with a 429 response. The error message usually tells you which limit you hit: "Rate limit reached for gpt-4o in organization org-abc on requests per minute (RPM): Limit 500, Used 500, Requested 1."

Your tier determines your limits. Free tier accounts get minimal access — often 3 requests per minute for GPT-4 models. Paid accounts start at Tier 1 and automatically upgrade as you spend more. Tier 5 accounts (those who've spent $1,000+) get substantially higher limits. You can check your current tier and exact limits at platform.openai.com/settings/organization/limits.

How to fix it

1. Check your current tier and limits

Go to platform.openai.com/settings/organization/limits. Look at your usage tier and the specific RPM/TPM numbers for each model. If you're on Free or Tier 1 with low limits, that's your problem right there.

2. Implement exponential backoff

Add retry logic to your code. When you get a 429 error, wait before trying again — start with 1 second, then 2, then 4, then 8. Most API libraries support this natively. In Python with the official OpenAI library, set `max_retries=3` when creating your client. The library handles the backoff automatically.

3. Batch your requests strategically

If you're sending 100 requests at once, you'll slam into RPM limits instantly. Space them out. Use a queue system or rate limiter in your code. Send requests at 90% of your limit to leave headroom — if your RPM is 500, aim for 450 per minute maximum.

4. Upgrade your tier

The fastest solution is spending more. Add $50 to your account and use it — that moves you from Tier 1 to Tier 2, roughly doubling your limits. At platform.openai.com/settings/organization/billing/overview, add credits and make sure you're actually using them. Tier upgrades happen automatically within 24 hours of hitting spending thresholds.

5. Switch models if appropriate

GPT-4o mini has much higher rate limits than GPT-4o. If your use case works with the smaller model, switching can eliminate 429 errors entirely. Check the limits page — you'll see GPT-4o mini often has 5-10x higher RPM allowances.

If that doesn't work

Contact OpenAI support to request a limit increase. Go to help.openai.com, click "Messages" in the bottom right, then "Send us a message." Choose "API" as the product and explain your specific use case. Include your organization ID (find it at platform.openai.com/settings/organization/general), which models you need higher limits for, and your expected request patterns.

Be specific about what you're building and why default limits don't work. "I need 10,000 RPM" won't help. "I'm building a customer service bot handling 200 concurrent users, each generating 2-3 requests per conversation" gives them actual context. For other common OpenAI API error 429 issues and authentication failures, check our dedicated troubleshooting guide.

Response times vary. Tier 4-5 accounts often hear back within 24-48 hours. Lower tiers might wait 3-5 days. If you have billing issues causing limits, see our guide on handling unexpected OpenAI charges.

Questions people actually ask

Q: Why am I getting 429 errors when I haven't used my account all day?

A: Rate limits reset every minute, but organization-wide limits affect all API keys under your account. If someone else on your team is hammering the API, you'll both hit the same limit.

Q: How long until my tier upgrades after I add money?

A: Usually within 24 hours of the calendar date you hit the spending threshold. Add $50 on Monday, hit Tier 2 limits by Tuesday. Check platform.openai.com/settings/organization/limits to confirm.

Q: Can I pay extra just for higher limits without upgrading tiers?

A: No. OpenAI's tier system is automatic based on total spending. The only path to higher limits is using the API more over time or requesting manual increases for specific business cases.

What to remember

  • Error 429 means you've exceeded requests per minute, tokens per minute, or daily request limits for your tier
  • Check platform.openai.com/settings/organization/limits to see your exact numbers and current tier
  • Implement exponential backoff in your code — most libraries do this automatically with retry settings
  • Upgrade tiers by spending more on the API — each tier roughly doubles your limits
  • Request manual increases through help.openai.com with specific use case details and your organization ID

---

*openai-support.com is an independent resource, not affiliated with OpenAI Inc.*

Related help