is now an ernd partner: 10% back on inference 
In this guide
What are the limits now?
Groq’s limits apply to your organization, not to each user or key. They’re counted in requests and tokens per minute and per day, and whichever one you hit first stops you. Cached tokens don’t count.
Here’s what Groq lists for one popular model, openai/gpt-oss-120b:
| Plan | Requests a minute | Requests a day | Tokens a minute | Tokens a day |
|---|---|---|---|---|
| Free | 30 | 1K | 8K | 200K |
| Developer | 1K | 500K | 250K | None listed |
Groq calls these a high level summary with exceptions. Your exact numbers are on the Limits page in your account settings in the Groq console. Some organizations also get separate input and output token limits. If yours does, hover over the TPM value there to see the split.
Your AI bill, with ernd
What could your AI bill earn?
Enter an amount from $0 to $100,000,000.
back a year at 4%
For US companies. No annual fee. Illustration based on AI vendor spend only. Rewards are points redeemable as cash back.
How do you move up?
- Upgrade to the Developer plan from the Groq console.
- Add a payment method. Groq takes Visa, Mastercard, Amex and Discover. A US bank account or SEPA debit works too.
- That’s it. The upgrade takes effect immediately.
Groq’s docs only list the two plans. Groq says higher limits are available for select workloads and enterprise use cases, so if Developer isn’t enough, talk to Groq about your traffic. Developer accounts get chat support, which is a good place to start.
What changes on your bill?
Nothing is charged when you upgrade. New accounts are billed progressively at first: an invoice goes out when lifetime usage first crosses $1, then $10, $100, $500 and $1,000. After $1,000 it’s one invoice a month. Customers billed in India have different thresholds.
Developer also unlocks spend limits. Set one, because a hard monthly cap is easier than finding out at month end. When it’s hit, every key returns a 400 with blocked_api_access. Here’s how to fix a Groq spend limit block.
How do you handle a 429 in the meantime?
When you go over a limit, Groq returns 429 Too Many Requests with a retry-after header in seconds. Back off for that long. The x-ratelimit-remaining-tokens header shows how much of your per-minute token budget is left, so you can slow down before you hit the wall.
Groq bills the card on file, so pick one that earns on it. ernd is a credit card for US companies with 4% back on AI vendor spend and 1.5% on everything else. Request the card at app.ernd.ai.
See how Groq compares with other vendors on our AI API rate limits page.
The ernd card
Keep capacity and card rewards separate.
The card earns rewards on AI spend. It does not bypass provider quotas, fix account restrictions or change provider prices.
Request your cardThe ernd card is a credit card for US companies, with its own partners in the AI stack. ernd isn't affiliated with any of the vendors on this page.
Common questions.
Does upgrading to Groq's Developer plan cost anything up front?
No. Groq doesn't charge at upgrade. You're billed for tokens at month end, or earlier when your usage first crosses $1, $10, $100, $500 and $1,000.
Do cached tokens count toward Groq rate limits?
No. Groq says cached tokens don't count toward your rate limits.
Are Groq rate limits per API key?
No. They apply to the whole organization, so every key and every teammate shares them.
Sources & methodology
Checked . Providers change prices and terms; the linked official page is the final word. This guide is published by ernd.
ernd is run by Provenance Research Inc.