What this crawler family is
Perplexity traffic is usually evaluated under the umbrella of AI answer engines, but site owners still need the narrower operational question: was this an autonomous crawler request or a user-triggered fetch?
provider
Perplexity Crawlers
Distinguish PerplexityBot from Perplexity user-triggered fetch traffic and decide whether your policy should preserve one while restricting the other.
Verification and source data
IPBot uses the source inventory below to decide whether a request is merely known or fully verified.
Swipe sideways to review every verification column.
| Crawler | Intent | Verification | CIDRs | Freshness |
|---|---|---|---|---|
| Perplexity-User
Official source
| user triggered | Known IP range | 4
IPv4 4 / IPv6 0 | 2025-10-17 |
| PerplexityBot
Official source
| assistant | Known IP range | 8
IPv4 8 / IPv6 0 | 2025-02-07 |
Whether to allow, block, or separate it
Keep PerplexityBot and Perplexity user-triggered fetches separate if your publishing or support flows care about live retrieval behavior.
robots.txt guidance
Use the documented Perplexity crawler controls directly rather than assuming a single AI block rule is precise enough.
Cloudflare and WAF guidance
Apply independent WAF or allowlist logic for PerplexityBot versus user-triggered traffic when the business impact differs.
Perplexity-User
Keep this separate from PerplexityBot if you want user-triggered retrieval to behave differently from autonomous crawling.
Useful as a separate allow/block policy for preview or live-answer fetches.
PerplexityBot
Use the documented Perplexity crawler token and keep its access rules explicit.
Treat as a distinct crawler family rather than bundling it into generic AI bot rules.
FAQ
Why should I separate PerplexityBot from Perplexity user-triggered fetches?
They serve different purposes. Autonomous crawler access and user-triggered retrieval do not always deserve the same allow, challenge, or block decision.
Does IPBot use a live runtime dashboard for these pages?
No. The public crawler pages are built from a generated catalog artifact produced from IPBot's crawler range snapshot at build time.