Rate Limits

When repeatedly calling the Atla API, you may encounter rate limiting errors. These are restrictions we impose via the API on the amount of times a user can access our models within a specific time period.

​

Why do we have rate limits?

Rate limits are essential to ensure the stability and reliability of our services. By imposing these restrictions, we can:

• Prevent Abuse: Protect our services from excessive usage that could be caused by automated scripts or malicious users.
• Maintain Performance: Help maintain optimal performance and responsiveness for all users by preventing overloading of the system.

​

How do these work?

We measure rate limits in two simple ways: RPM (requests per minute) and RPD (requests per day). Rate limits could be hit across any of the two options depending on which one occurs first.

​

Tiers

See our rate limit tiers below.