is now an ernd partner: 10% back on inference 
In this guide
What are the limits on each key?
Cohere has two kinds of API key. Trial keys are free and limited. Production keys are paid and much less limited. A few chat models from Cohere’s table:
| Model | Trial key | Production key |
|---|---|---|
| Command A | 20 req/min | 500 req/min |
| Command R+ | 20 req/min | 500 req/min |
| Command R7B | 20 req/min | 500 req/min |
| Command A Reasoning | 20 req/min | Contact sales |
Other endpoints have their own numbers. Rerank goes from 10 to 1,000 requests a minute. Embed stays at 2,000 inputs a minute on both kinds of key.
Trial keys also stop at 1,000 API calls a month. So do production keys on the newer chat variants.
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 get a production key?
- Go to the API keys page in the Cohere dashboard.
- Start the go-live steps. You accept Cohere’s SaaS agreement and terms of service, and confirm your team has read the model limitations, model cards and data statement.
- Say whether your use falls under any of the sensitive use cases in Cohere’s usage guidelines.
- Create the production key. If you didn’t flag a sensitive use, it works right away.
If you did flag one, the key may stay at trial limits until Cohere’s safety team reviews it. Cohere says that takes no longer than 72 business hours.
What if production limits aren’t enough?
Email support@cohere.com. That’s the address Cohere gives for rate limit increases. For the newer variants that stay at trial limits, it’s sales@cohere.com.
Which error are you seeing?
A 429 is the rate limit. Cohere’s examples include You are past the per minute request limit, please wait and try again later. and, on a trial key, trial token rate limit exceeded, limit is 100000 tokens per minute. If you see the trial wording, a production key is the fix.
A 402 is billing, not rate limits. Here’s how to fix a Cohere 402.
Production keys are billed to the card in your dashboard. 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 Cohere 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.
How long does Cohere's sensitive use case review take?
No longer than 72 business hours, according to Cohere. Until it's done, your production key may be rate limited like a trial key.
Do production keys raise limits on every Cohere model?
No. Cohere says production keys work like trial keys on newer variants such as Command A Reasoning, and its table lists Command A+, Command A Translate and Command A Vision as contact sales too. Cohere asks you to email sales@cohere.com to use them in production.
Who do I contact for a Cohere rate limit increase?
Cohere's docs point to support@cohere.com for rate limit increases.
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.