Checked 18 Sept 2026. HTTP 403
Perplexity Pro
A grounded answer costs real money on both sides — the search call and the long retrieved context — and the estimated multiple stays in low single digits.
- Vendor
- Perplexity AI
- Category
- Search
- Tier
- Pro
- Price / month
- $20
- Est. API cost
- $5.25
- Est. markup
- 3.8×
- Price checked
- 11 Aug 2026
- Verification
- verified
What the price actually is
$20/month billed monthly; an annual rate is advertised lower. Pro is frequently bundled free for a year through carrier and hardware promotions, which many subscribers are on.
The usage this is priced against
A regular research user: 100 searches a month, each running one web search and then answering over roughly 12,000 tokens of retrieved page text.
Someone using it half as much sees double the multiple. The assumption is the argument — if you disagree with it, the number below is not about you. Change it in the calculator.
The cost math, in full
Estimated monthly inference cost
Web search API at $6.00 / 1,000 search queries · one search API call per question
Mid-tier frontier text model at $3.00 / 1M input tokens · grounded answer over retrieved passages, with citations
Mid-tier frontier text model at $15.00 / 1M output tokens
Rates come from a dated rate card of representative published API prices, not from the vendor. Nobody outside these companies knows what they actually pay; volume discounts and in-house serving both push real costs below these figures, which makes every multiple here a floor rather than a ceiling.
Why this verdict, not the number
Retrieval-augmented search is one of the few consumer AI products where the non-model costs are genuinely large: every question pays for crawling or a search API before a token is generated, and the retrieved context makes the input side of the model bill unusually heavy. A single-digit multiple over that combined cost is not where the site's attention belongs.
What the price buys besides tokens
- Data pipeline
- Orchestration
- UX craft
The index and the freshness are the asset. Query rewriting, source selection, deduplication and citation alignment are all work that sits between the search call and the answer, and all of it degrades quietly rather than failing loudly when done badly.
What you lose if you leave
Citations that line up with the sentences they support, a retrieval layer tuned over a large query volume, and freshness on fast-moving topics.
The honest cheaper path
A search API key plus a model call gets you a working clone in an afternoon. Getting citations to actually match the claims is the part that takes months.
Recompute it for yourself
Recompute this for your own usage
Scaling assumes your usage has the same shape as ours, just more or less of it. If your mix is different — far more output than input, say — the estimate drifts. It is an estimate either way.
Where every number came from
- pricingPerplexity Pro
- rate-cardBrave Search API pricing
- rateRate card entry: web-search-api
- rateRate card entry: frontier-mid
Price recorded 11 Aug 2026 · entry last reviewed 11 Aug 2026. Think something here is wrong? File a correction — we publish them, including the ones that embarrass us.