Head to head
unlob vs Perplexity
Perplexity sells raw search over its own index at $1 per 1,000 with Fast Search, and answers through its Agent API. unlob sells what a result list leaves out: independence, coverage receipts and a status.
Verified 7 October 2026·Figures from vendors' own pages
On the headline rate that is 2.5× — $1 per 1,000 for Perplexity’s Search API with Fast Search, its cheapest comparable mode, against $0.40per 1,000 searches. Other modes cost more; the pricing detail below lists them. Both figures are computed from the published plan tables rather than typed into this page, so they follow the vendors rather than drifting from them.
Price
- unlob$0.40
- Perplexity$1
| Provider | Own index? | $ / 1,000 searches | Source |
|---|---|---|---|
Priced: Search API with Fast Search. The Search API is $5 per 1,000 requests, or $1 with Fast Search, with no token charges (up to 20 results, 10 by default). Sonar Chat Completions support ended on 27 September 2026 and requests are moving to the Agent API; Sonar charged $5 to $12 per 1,000 by search context size plus $1/$1 per million tokens, and Sonar Pro $6 to $14 plus $3/$15. | Yes Perplexity index | $1 | Perplexity API pricing |
$29 per month for 50,000 credits, and a search is one credit. | Yes Our own crawl and index | $0.58 | This page |
$199 per month for 500,000 credits, and a search is one credit. | Yes Our own crawl and index | $0.40 | This page |
When Perplexity is the better choice
A comparison page that cannot name a scenario where the competitor wins is an advertisement. Here are the real ones.
Choose Perplexity
- You want a plain, fresh result list from a large index at $1 per 1,000, and your agent does its own evidence handling.
- You want an answer, not evidence, and do not need to control how it was assembled — the Agent API does retrieval, ranking and synthesis in one call.
- Perplexity’s index and freshness on news-shaped queries are strong.
Choose unlob
- You need to know which results are independent of each other, not just which ones ranked highest — five articles repeating one press release are one origin here.
- You need a record of what was searched, what was not and what is missing, so an empty or thin result means something.
- You want a deterministic status — sufficient, insufficient, stale, partial or empty — so your model can abstain instead of answering from whatever came back.
- Graph traversal and coverage auditing.
The substantive differences
| Perplexity | unlob | |
|---|---|---|
| What you get back | A ranked result list from the Search API, or a written answer with citations from the Agent API. | Passages with provenance, independent origins and a coverage receipt from ground, plus ranked passages from search. Never a generated answer. |
| Pricing | Search API $1 per 1,000 with Fast Search or $5 standard, no token charges. Answers through the Agent API add model token charges. | $0.40 per 1,000 searches (a search is one credit), no token component; ground is 5 credits. |
| Model choice | Your own model over Search API results, or Perplexity’s models through the Agent API. | Whichever model you like, or none. |
Sources
Every figure on this page was read from the vendor's own pricing page or funding announcement on the date shown.
- Perplexity: Perplexity API pricingread 7 October 2026
- Perplexity: Perplexity Fast Searchread 7 October 2026
- Perplexity: Migrating from Sonar to the Agent APIread 7 October 2026
Frequently asked questions
Sonar Chat Completions ended — what should I move to?
Perplexity is reformulating synchronous Sonar requests onto its Agent API and recommends it for new projects; asynchronous Sonar requests are no longer supported. If you are re-integrating anyway, decide first whether you want the vendor to synthesise the answer at all. If your own model writes the answer, what you need from the search layer is evidence your model can weigh — which is the part unlob is built around.
Is Perplexity’s Fast Search cheaper than unlob?
No — $1 per 1,000 against $0.40 — but the gap is 2.5×, which is small enough that price should not decide it. Decide on what your agent needs back: a ranked list, or evidence with independence and coverage stated.
Check the claims yourself
The free tier is 10,000 credits a month with no card. That is enough to test coverage, latency and result quality against your own queries — which is a better comparison than this page.