Unreachable RPM values on Agent API Rate Limits docs table

Bug Reports
,
1

Pre-flight checklist

  • This is about the Perplexity API, not the Perplexity app, Computer, Comet, or web UI.
  • I have removed API keys, secrets, and private data.
  • I have included enough information for someone else to reproduce the issue.

Bug type

Bug type

  • Other API bug
    • API docs issue

Affected API area

  • Agent API

Summary

The Agent API Rate Limits says

The Agent API applies two independent limits to each request: the API-wide QPS limit and the Agent model’s requests-per-minute (RPM) limit. A request must be within both limits. The RPM value is not derived from the QPS value

For Tier 4 how can we ever reach 4,000/min if the 33 QPS is being enforced? 33*60 = 1980

Same question for Tier 5.

Expected behavior

QPS and RPM values being congruent.

Actual behavior

RPM is in theory unreachable.

2

@dev You’re right: 33 requests per second cannot sustain 4,000 requests per minute.

We updated our documentation: Rate Limits & Usage Tiers - Perplexity

Agent API does not have the separate per-model RPM. Thank you for flagging it.

Andrew

3

Thanks for your answer, Andrew.

Is there a strong reason for cutting down the Rate Limit in less than half for production applications? We’re noticing more rate limit hits since the migration. Is there a way to request for increase back to ~67 QPS?
pls let me know if I should open a separate ticket; also contacted API support email

4

Hello @dev,

Please fill out a rate limit increase request for the team to review.

Andrew

5

Thanks, @andrewmadson-pplx , Albert already responded my email.