By use case · Tavily pricing

Tavily pricing: API credits, cost per search, extract and crawl

The short answer

Tavily sells monthly API credits. The Researcher tier is free with 1,000 credits a month, then Project is $30 for 4,000 credits, Bootstrap $100 for 15,000, Startup $220 for 38,000 and Growth $500 for 100,000, with pay-as-you-go at $0.008 a credit and Enterprise quoted. One credit buys one basic search, or five extracted URLs, or ten mapped pages. Crawl bills as mapping plus extraction, so ten crawled pages with basic extraction is three credits. That puts a basic crawl at $1.50 per 1,000 pages on Growth and $2.25 on Project. Two rules move the real number: the volume discount is unusually shallow, and extract bills per batch of five, so calling it one URL at a time costs up to five times as much. Every figure here was read from the Tavily pricing page and API documentation on 16 September 2026.

Clean markdown & JSON · JavaScript rendered · robots.txt respected

Extraction demo

Endpoint: POST /v1/extract · paste a URL, or pick a sample below

POST

Try:

Hit Extract to turn this page into clean, LLM-ready data.

robots.txt respected · public data only ·

Markdown · JSON · structured fields, from one API call. Crawling, rendering and extracting ... Samples are recorded. Your own URL is crawled live.

Tavily publishes its plan ladder and its credit table, and both are easy to find. What neither page does is convert the two into a cost per page, which is the number you actually need before you commit a budget.

Start with the credit table. A basic search is 1 credit and an advanced search is 2. Extract is 1 credit per five successful URLs on basic depth and 2 per five on advanced. Map is 1 credit per ten pages returned, or 2 per ten when you pass instructions. Crawl is not priced separately at all: it bills as the mapping cost plus the extraction cost, which Tavily illustrates with ten pages at basic depth coming to three credits. Failed extracts and failed map requests are not charged, which is a genuinely good rule and one several rivals do not follow.

Now the ladder. Project works out at $0.0075 a credit, Bootstrap $0.0067, Startup $0.0058 and Growth $0.0050. Pay-as-you-go is $0.008. That range is narrow by the standards of this category, and it has a consequence most buyers miss, which the first card below spells out.

We sell a crawl API, so treat the last section with the skepticism it deserves. Tavily is the right purchase for a large class of teams and we are not a substitute for it: its core product is live web search for agents, it has a free tier, and we offer neither.

If you are weighing the two product shapes rather than the price list, crawl API against search API sets them side by side, and Tavily alternatives compared on cost and output covers the rest of the field. Teams already batching extract calls should read when batch extraction stops being the cheap option, and our web scraping API pricing breakdown puts the whole category on one page.

CRAWL RENDER JS EXTRACT MARKDOWN JSON

Any URL in LLM-ready data out

robots.txt respected public data only

Why it works

Three things the Tavily pricing page does not work out for you

The $30 plan saves you $2

Project is $30 for 4,000 credits, which is $0.0075 each. The same 4,000 credits on pay-as-you-go cost $32.00 at $0.008. The commitment buys a 6.25 percent discount. Bootstrap saves 17 percent, Startup 28 percent and Growth 38 percent, so the ladder only starts paying for itself well above the entry tier.

Extract bills in fives

The documented rate is 1 credit per five successful extractions, and the endpoint takes 1 to 20 URLs in a single call. Tavily does not publish how a partial batch rounds, so budget for rounding up: a one-URL call costs the same credit as a five-URL call. An agent extracting URLs as it finds them can pay five times a batching pipeline for identical work.

Crawl is two meters, not one

There is no crawl rate. Tavily charges the map cost plus the extract cost, so ten pages at basic depth is 1 mapping credit plus 2 extraction credits, or 3 in total. Ten pages at advanced depth is 1 plus 4, or 5. That is 0.3 and 0.5 credits a page, which is where the per-1,000 figures in the table come from.

What it handles

Any URL in, clean structured data out

Point ClawEngine at a public page and it crawls, renders the JavaScript and extracts clean markdown or typed JSON in one call. Define a schema for structured fields, and respect robots.txt and Terms of Service by default.

Every published Tavily plan and rate, verified 16 September 2026
What one credit buys on search, extract, map and crawl
Cost per 1,000 searches, extracted URLs and crawled pages on every tier
Where pay-as-you-go beats a subscription and where it does not
Published rate limits for development and production keys
Where Tavily is the right purchase and a flat per-page crawl API is not

POST /v1/extract extraction result

200 · JSON

{
  "url": "https://example.com/products/atlas",
  "title": "Atlas Field Notebook",
  "markdown": "# Atlas Field Notebook\n\nDurable...",
  "data": {
    "name": "Atlas Field Notebook",
    "price": 24.00,
    "currency": "USD",
    "rating": 4.7
  },
  "links": [ "/products", "/cart" ],
  "metadata": { "rendered": true }
}

JS rendered · boilerplate stripped ✓ robots.txt respected

Head to head

Tavily pricing by tier, with a flat per-page crawl API in the third column

Every Tavily figure below was read from its pricing page, help centre and API documentation on 16 September 2026, never from a review aggregator. We put ourselves third and we lose several rows outright.

Swipe to compare all columns →

What you are comparing	Tavily plans	Tavily pay-as-you-go	ClawEngine
What the meter counts	Credits, at a different rate per endpoint	The same credits, bought one at a time	Pages. One page of plan allowance per page returned
Entry price	Free forever with 1,000 credits a month, then $30 on Project	No commitment at $0.008 a credit	$39 a month on Hobby. No free plan
Published tiers	Researcher free 1,000, Project $30 for 4,000, Bootstrap $100 for 15,000, Startup $220 for 38,000, Growth $500 for 100,000, Enterprise quoted	Any volume, billed per credit used	Hobby $39 for 50,000 pages, Startup $99 for 250,000, Scale $399 for 1,500,000, Enterprise custom
Price per credit	$0.0075 Project, $0.0067 Bootstrap, $0.0058 Startup, $0.0050 Growth	$0.008 flat	Not applicable. We meter pages, not credits
Cost per 1,000 basic searches	$7.50 Project, $6.67 Bootstrap, $5.79 Startup, $5.00 Growth	$8.00	Not offered. We do not run web search at any price
Cost per 1,000 extracted URLs, basic depth	$1.50 Project, $1.33 Bootstrap, $1.16 Startup, $1.00 Growth	$1.60	$0.78 Hobby, $0.40 Startup, $0.27 Scale
Cost per 1,000 crawled pages, basic depth	$2.25 Project, $2.00 Bootstrap, $1.74 Startup, $1.50 Growth	$2.40	$0.78 Hobby, $0.40 Startup, $0.27 Scale
Cost per 1,000 crawled pages, advanced depth	$3.75 Project, $3.33 Bootstrap, $2.89 Startup, $2.50 Growth	$4.00	The same as basic. There is one depth and one rate
Cost per 1,000 mapped URLs	$0.75 Project, $0.67 Bootstrap, $0.58 Startup, $0.50 Growth	$0.80	Discovery is included in the crawl allowance
Structured JSON against your own schema	Not a published endpoint. Content comes back as markdown or text for a model to read	The same	Included at the same per-page rate
Billing on failures	Failed extracts and failed map requests are not charged	The same rule	Pages successfully returned to you
Batch size on extract	1 to 20 URLs per call, billed per five successful extractions	The same	Per page, so batch size does not change the bill
Rate limit per minute	100 on a development key, 1,000 on production. Crawl is 100 on both	The same limits apply	Concurrency scales with tier
Live web search for agents	The core product, with ranked results and content included	The same	Not offered. You supply the URLs or a seed to crawl from
Best for	Agents that need to find pages nobody named in advance, at steady monthly volume	Spiky or seasonal workloads under a few thousand credits a month	Pipelines crawling sites you already know, where output has to arrive as typed fields

Why ClawEngine

One API that crawls, renders and extracts

Not a raw HTML dump, not a headless browser fleet to run, and not a brittle parser to maintain. One call crawls a public page, renders its JavaScript and returns clean markdown or typed JSON, built for RAG pipelines and AI agents.

LLM-ready output

Clean markdown or typed JSON with the boilerplate stripped, so the data drops straight into a vector store, a prompt or an agent without a cleanup step.

JavaScript rendered

Each page loads in a real browser environment before extraction, so single-page apps and client-rendered content come back complete, not as an empty shell.

Compliance-first

ClawEngine works on public, permitted data only. It respects robots.txt and site Terms of Service and honors crawl-delay, so responsible scraping is the default.

Code examples

Crawling known sites without paying two meters

Tavily charges a crawl as mapping plus extraction because discovery is a separate product for it. If the sites are already on your list, that second meter is buying you nothing. Send a seed URL, a page budget and a schema, and the whole job bills as pages.

curl Crawl a known site into typed fields (POST /v1/crawl)

curl https://clawengine.ai/v1/crawl \
  -H "Authorization: Bearer $CLAWENGINE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/docs",
    "max_pages": 500,
    "include_paths": ["/docs/*"],
    "format": "json",
    "schema": {
      "title": "string",
      "section": "string",
      "body_markdown": "string"
    }
  }'

# Discovery and extraction are one meter: 500 pages of plan allowance.
# No separate map credit, no batch-of-five rounding, no format surcharge.

Tavily pricing: the questions buyers ask

How much does Tavily cost?

Tavily has a free Researcher tier with 1,000 API credits a month and no card required. Paid plans are Project at $30 a month for 4,000 credits, Bootstrap at $100 for 15,000, Startup at $220 for 38,000 and Growth at $500 for 100,000. Pay-as-you-go is $0.008 a credit with no commitment, and Enterprise is quoted with custom rate limits and SLAs.

What is a Tavily API credit?

One credit is the unit every endpoint is billed in. A basic search costs 1 credit and an advanced search 2. Extract costs 1 credit per five successful URLs at basic depth and 2 per five at advanced. Map costs 1 credit per ten pages returned, or 2 per ten with instructions. Crawl has no separate rate and bills as mapping plus extraction.

How much does Tavily cost per search?

A basic search is 1 credit, which is $0.0075 on Project, $0.0067 on Bootstrap, $0.0058 on Startup, $0.0050 on Growth and $0.008 on pay-as-you-go. Per 1,000 basic searches that is $7.50, $6.67, $5.79, $5.00 and $8.00. An advanced search is 2 credits, so double every one of those figures.

How much does Tavily Extract cost?

Basic extract is 1 credit per five successful URLs, so 1,000 URLs is 200 credits: $1.50 on Project, $1.33 on Bootstrap, $1.16 on Startup, $1.00 on Growth and $1.60 on pay-as-you-go. Advanced extract doubles that to 400 credits, or $3.00 down to $2.00 across the same tiers. Failed extractions are not billed.

How much does Tavily Crawl cost per page?

Crawl bills as mapping plus extraction. Ten pages at basic depth is 3 credits and ten at advanced depth is 5, so 0.3 and 0.5 credits a page. Per 1,000 crawled pages at basic depth that is $2.25 on Project, $2.00 on Bootstrap, $1.74 on Startup, $1.50 on Growth and $2.40 on pay-as-you-go. Advanced depth costs roughly 67 percent more.

Is Tavily pay-as-you-go cheaper than a plan?

Below about 4,000 credits a month it is close enough not to matter. Pay-as-you-go is $0.008 a credit and Project works out at $0.0075, so committing $30 a month saves $2.00 on the same 4,000 credits. The gap widens higher up: Bootstrap saves 17 percent against pay-as-you-go, Startup 28 percent and Growth 38 percent.

What are the Tavily rate limits?

Search, extract and map allow 100 requests a minute on a development key and 1,000 a minute on a production key. Crawl is capped at 100 a minute on both. The research endpoint allows 20 a minute on both, and the usage endpoint allows 10 requests per 10 minutes. Enterprise rate limits are set by contract rather than published.

How many URLs can Tavily Extract take in one request?

Between 1 and 20 URLs per request, with one extract_depth and one output format applied to the whole batch. Because billing is documented per five successful extractions, a full 20-URL call is 4 credits at basic depth. Splitting the same 20 URLs across 20 single-URL calls should be budgeted at 20 credits, so batching is the single biggest cost lever on this endpoint.

Does Tavily have a free tier?

Yes. The Researcher plan is free forever with 1,000 API credits a month and no credit card, which is 1,000 basic searches, or 5,000 extracted URLs, or roughly 3,300 crawled pages at basic depth. There is also a separate free program for students. This is a real advantage over vendors, ourselves included, that start at a paid tier.

Does Tavily offer volume discounts?

It does, but a shallow one by category standards. The published ladder runs from $0.0075 a credit on Project to $0.0050 on Growth, a 1.5x improvement across a 25x jump in volume. Vendors that meter plain pages typically compress five-fold or more over the same range, which is a fair reflection of Tavily running live search rather than fetching a URL you supplied.

Good questions

Questions about Tavily pricing

Not for crawling pages you already know about. A basic crawl costs 0.3 credits a page, which is $1.50 per 1,000 pages on Growth and $2.25 on Project, against $0.40 per 1,000 on our Startup tier and $0.78 on Hobby. Tavily is priced like a search product, because that is what it is, and its search endpoint has no equivalent here at any price.

No, not on extract or map. The credits documentation states plainly that a failed URL extraction is not charged and a failed map request is not charged. That is worth real money on a stale URL list, and it is a better rule than several competitors apply, where any returned document including a 403 or 404 bills at full rate.

Read every web scraping question

Explore more

More ways to turn the web into data with ClawEngine

See every use case See pricing Back to the web scraping API

Stop wrangling raw HTML. Get LLM-ready data.

Point ClawEngine at a public page and one call crawls, renders the JavaScript and extracts clean markdown or typed JSON, ready for your RAG pipeline or AI agent. Public, permitted data only.

See pricing

Crawl · render JS · extract markdown & JSON · robots.txt respected, public data only