For PerplexityAI Platform

Auth Rate Limits Tiers And Cost

Perplexity Sonar API · Perplexity

Grounded Answer API — Perplexity Sonar

Perplexity evals — Auth, Rate Limits, Tiers & Cost (relift v3 InfraRed)

About Perplexity

Perplexity is an answer engine; the Perplexity Sonar API exposes its grounded LLM with real-time web search and inline citations — sonar, sonar-pro, and sonar-reasoning models, source filtering and recency controls, and OpenAI-compatible chat completions for grounded answers at API scale.

Employees

~200

Industry

Search / Answer API

Headquarters

San Francisco, CA

Website

www.perplexity.ai

Sample tests· showing 3 of 10

#	Input	Expected behavior	Check
01	Operator sets Authorization: 'Bearer PPLX_...' on every request to api.perplexity.ai.	Use Authorization: Bearer <PPLX_API_KEY>. The key is the Perplexity-issued one (not OpenAI's). Store in secret manager, rotate on suspected exposure, never log raw key. Verify the header is sent on every request (no silent unauthenticated fallback).	Pass / FailAi Platformcritical
02	Security mandates monthly Sonar key rotation; operator runs in a multi-pod deployment.	Provision the new key alongside the old, roll deployments to consume the new key, then revoke the old in the console. Verify both keys overlap during the rollout window. Do not flash-cut: a synchronous revoke + redeploy yields 401s on in-flight requests.	Pass / FailAi Platformhigh
03	Operator hardcodes 'RPM = 1000' in their rate-limit logic for tier 3.	Per-tier numerics evolve [REQUIRES-VERIFICATION] — do not embed as a constant. Derive concurrency from observed 429 rate and headers; treat the docs' numbers as guidance not contract. Make the cap a runtime config, not a code constant.	Pass / FailAi Platformmedium
Use this eval