Trial limitations

Trial limitations

Hey! Hope you're staying safe!!
I'm trying to benchmark a model deployed (and trained) in Datarobot by hitting it's real-time prediction endpoint with up to 120 requests per second. We're trying to determine if the auto-scaling capabilities of the deployments are good enough for our use cases.

Requests are sent like:

  1. Seconds 0 to 60: 2 requests per second.
  2. Seconds 61 to 120: 10 requests per second.
  3. Seconds 121 to 180: 120 requests per second.
  4. Seconds 181 to 240: 70 requests per second.
  5. Seconds 241 to 300: 5 requests per second.

At about 125 seconds into the benchmark, I get a limit error: "You have exceeded your limit on total modeling API requests. Try again in 1 seconds.".


I tried using the "contact us" button to ask for help, but they directed me to this forum. Does someone here know what the rate limits for real-time predictions on the free trial are? Is there a way to extend these limits?



Labels (1)
3 Replies

Of course Sebastian!


I think the best course of action would be to contact our sales team in order to test the full functionality. In the app, click on the DataRobot logo in the top left to navigate to AI Platform Home. From here, you can click Get in Touch > Contact sales. They should be able to help you out.





Hey Taylor, thanks for the insight.


Is there any way to pay to enable the full functionality, but for a short while? Say a month or a week.


In particular, we'd like to test the increased throughput and the ability to deploy custom models, which is currently not present in the free trial.



Hi  @Sebastian Correa ,


Thanks for the question and hope you're staying safe as well!


The rate limit for a free trial account is 50 requests per second, so it makes sense that you hit this error. I'd try reducing the rate of your requests. Note that this limit only applies to the trial accounts and cannot be extended. For more info about these accounts, feel free to take a look at