How to raise your Mistral rate limits

Where do you see your limits?

In the Admin Panel, under API, then Limits. Completion limits are listed per model, as tokens per minute and requests per second. Audio and document OCR have their own limits on the same page.

How do you ask for more?

Contact Mistral support. Mistral has no self-serve tier to climb, so the request is the whole process. A complete one saves you a round of back and forth. Mistral asks for:

1. The models or APIs you plan to use.
2. Your target requests per second.
3. Your expected tokens per minute and tokens per month.
4. A short description of your use case.

Use real numbers from the usage dashboard if you have them. It breaks usage down by model, input and output tokens, and workspace, which covers most of what support will ask.

Is it a rate limit or a spending limit?

Check before you write to support. The two are separate, and the fix is different.

The organization spending limit caps API and Vibe consumption for the month. If you reach it, API access can be suspended until the next month or until an admin raises it. Failed invoice payments can affect access as well. An admin sets the limit under Subscriptions, then Billing.

Each workspace can also have its own cap, which can’t exceed the organization’s. It’s set under Administration, then Workspaces, on the workspace’s Settings tab. If one team is blocked and the others aren’t, look there first.

Free mode has its own limits. Pay-as-you-go extends usage past the included monthly amount and draws from your prepaid credits.

See how Mistral compares with other vendors on our AI API rate limits page.