Skip to the page
Get started free

Parallel pricing

$1 to $5 per 1,000 searches, and $1 on SpecStream.

Parallel's Search API costs $1 per 1,000 requests in turbo and fast mode and $5 in basic and advanced, and a request with no mode runs advanced. Results past the first 10 cost $1 per 1,000, and Extract costs $1 per 1,000 URLs.

SpecStream is the live web data engine for teams and agents. It answers Parallel's own /v1/search call at $1 per 1,000 searches in turbo, fast and basic mode, reads pages at $0.80 per 1,000 and delivers datasets at 5¢ to $1 a row.

Get an API key

New accounts get $10 of credit, no card. It covers 10,000 searches.

Every Parallel API price, and what SpecStream charges for the same job

SpecStreamParallel
Web search, per 1,000 calls$1 per 1,000 searches of up to 10 results on Parallel's own call, https://api.specstream.ai/parallel/v1/search, in turbo, fast and basic mode. $5 in advanced$1 per 1,000 requests in turbo and fast, $5 in basic and advanced. A request with no mode runs advanced, at $5
More than 10 resultsUp to 100 a call. 11 to 20 results cost $0.0022 in fast, and each started block of 10 more adds $0.0012$1 per 1,000 results beyond the first 10, so 20 results cost $0.011 in fast and $0.015 in advanced. Public modes cap results at 20
Page textbasic reads each page and returns up to about 3,500 characters of it in the $0.001 search. The Extract API reads up to 20 URLs a call at $0.80 per 1,000 pages, and a page with no text is not chargedExtract API $1 per 1,000 URLs. Excerpt length does not change the search price
A cited answer$5 per 1,000 searches. The reply carries answer with its page and read time, rows with the facts asked for, and the results readResponses API per 1,000 successful requests: $10 on low, $50 on medium, the default, and $250 on high. Decisions API $1 per 1,000 requests of up to 100 questions
Research and listsDeep Search: every match on the live web as a row, with a source URL and read time on every value, at 5¢, 15¢ or $1 a delivered row. No fee per run, and the most a run can cost is quoted before it startsTask API $5 (lite) to $2,400 (ultra8x) per 1,000 successful runs. FindAll $0.25 a run plus $0.03 a match on base, $10 a run plus $1 a match on pro, with enrichments billed per match on top
Watching for changeA saved request reruns hourly to yearly. Each row comes back labeled changed, unchanged or no longer matching. A changed or new row costs half its rate, an unchanged row 1¢Monitor API $3 (lite) or $10 (base) per 1,000 checks. Each execution counts as one check
Free use$10 of credit on a new account, no card: 10,000 searches through Parallel's call, 2,000 with the answer, or 12,500 pages$5 free monthly with a card on file, plus up to $80 at signup. A free Search MCP with variable limits, enough for personal use
PlansThe same rate from the first search, with no plan to buy. Credit you buy never expiresBuild is pay as you go. Scale is $1,000 a month for $1,000 in credits, or $10,000 a year for $12,000. Unused monthly credits expire at the end of each paid month
Speed limitNo limit per key and no daily limit. A busy call answers 429 with Retry-After and is not chargedSearch and Extract: 600 requests a minute on Build, 1,500 on Scale, excluding Turbo
Zero data retentionOn any API key: set zero_retention and a search keeps no request text and no answer after the replyOn Scale as Hybrid ZDR once a signed agreement is in place, and on Enterprise by contract

What the Parallel price list means for your bill

The default mode is the $5 mode

Parallel's headline price is $1 per 1,000 requests, but a request that sets no mode runs advanced at $5. At 100,000 searches a month that is $500. basic mode costs the same $500 on Parallel, and $100 on SpecStream through the same call, with each page read.

Ask for 20 results and Parallel adds $1 per 1,000 results past the first 10: $0.011 a call in fast, $1,100 for 100,000 searches. On SpecStream the same 20 results cost $0.0022 a call, $220 a month.

Pages at $0.80 per 1,000

Parallel's Extract API costs $1 per 1,000 URLs. The SpecStream Extract API costs $0.80 per 1,000 pages, reads up to 20 URLs a call, and charges nothing for a page that is blocked, missing or empty. A million pages cost $800, against $1,000.

Rows billed when delivered, with no run fee

Parallel bills FindAll with a fixed fee per run plus a charge per match, and adds each enrichment's Task API price per match. A run that finds little still pays its fee.

Deep Search bills only delivered rows, at 5¢, 15¢ or $1 each, plus 5¢ for a person's email and 15¢ for a phone. Duplicates and rows a page contradicts are not charged, a run that finds nothing costs nothing, and nothing runs until you approve its maximum.

The most accurate web search API

The SpecStream Search API scores 97.9 F-score on all 1,000 official SimpleQA Verified questions, 978 of them correct, and 64.0 item F1 and 43.4 row F1 on all 200 official WideSearch tasks. Every score and its method are on the benchmarks page.

Other search API prices next to Parallel

SpecStream

The live web data engine for teams and agents. The Search API costs $1 per 1,000 searches of up to 10 results on the fast tier and every drop-in address, and $5 with the answer. The Extract API reads pages at $0.80 per 1,000, and Deep Search delivers full datasets at 5¢, 15¢ or $1 a row. New accounts get $10 of credit, no card.

Brave Search API

A search API on Brave's own index, sold as prepaid credits. The Search plan costs $5 per 1,000 requests with $5 in free credits every month. SpecStream answers Brave's own call at $1 per 1,000.

Exa

Search costs $4 per 1,000 requests on lite and $7 on auto and fast, and page contents cost $1 per 1,000 pages. SpecStream answers Exa's own call at $1 per 1,000 and reads pages at $0.80 per 1,000.

Tavily

Pay as you go is $0.008 per credit. A basic search costs 1 credit and an advanced search 2, so 1,000 searches cost $8 to $16, with 1,000 free credits every month. SpecStream answers Tavily's own call at $1 per 1,000.

You.com

Web Search costs $5 per 1,000 calls, and the Contents API $1 per 1,000 pages. SpecStream answers You.com's own call at $1 per 1,000 searches.

How SpecStream scores

The same numbers as the benchmarks page, from the official tasks and scorers.

SpecStreamParallelExaTavilyPerplexity
Web researchWideSearch, GPT-5.6 Luna60.945.553.047.947.0
Factual accuracySimpleQA Verified, GPT-5.6 Luna9694919494
Factual accuracySimpleQA Verified, GPT-5.6 Sol9997919295

How to move your Parallel calls to SpecStream

Change one line. https://api.specstream.ai/parallel/v1/search answers the call your code sends to https://api.parallel.ai/v1/search today, in Parallel's JSON, so your parser stays as it is. Send a SpecStream key in x-api-key. Each call of up to 10 results costs $0.001 in turbo, fast and basic mode.

# Before
curl -s https://api.parallel.ai/v1/search \
  -H "x-api-key: $VENDOR_API_KEY" -H "Content-Type: application/json" \
  -d '{"search_queries": ["Mistral AI founders"], "mode": "fast"}'

# After
curl -s https://api.specstream.ai/parallel/v1/search \
  -H "x-api-key: $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"search_queries": ["Mistral AI founders"], "mode": "fast"}'

In the parallel-web SDK, set base_url="https://api.specstream.ai/parallel". search_queries, objective, mode, max_chars_total, session_id and advanced_settings work as before, and the reply keeps results[] with url, title, publish_date and excerpts. Every field is in the Parallel migration guide.

For the answer in place of the Responses API, call SpecStream's own Search API at $5 per 1,000: GET https://api.specstream.ai/v1/web-search with Authorization: Bearer <key>, and the question in q in everyday words.

curl -G https://api.specstream.ai/v1/web-search \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" \
  --data-urlencode "q=Who leads engineering at Mistral AI, and when did they join?" \
  --data-urlencode "passages=true"

On MCP, your agent calls search_web with {"q": ...} and read_pages with {"urls": [...]} at https://api.specstream.ai/mcp. The SDKs call it web_search(q=...) in Python and webSearch({ q }) in TypeScript.

Get an API key

Frequently asked questions

Something we didn't answer? Talk to a human

How much does Parallel cost?
Search costs $1 to $5 per 1,000 requests by mode, with $1 per 1,000 results past the first 10. Extract and Decisions cost $1 per 1,000, Task $5 to $2,400 per 1,000 runs, Responses $10 to $250 per 1,000 requests and Monitor $3 to $10 per 1,000 checks. FindAll is a fee per run plus a charge per match.
How much does the Parallel Search API cost per request?
$0.001 in turbo and fast mode and $0.005 in basic and advanced, for 10 results. A request with no mode runs advanced, so it costs $0.005. On SpecStream, Parallel's own call costs $0.001 in every mode but advanced.
Is Parallel free?
Parallel gives $5 free monthly with a card on file, plus up to $80 at signup, and a free hosted Search MCP for personal use. SpecStream gives a new account $10 of credit with no card, which covers 10,000 searches.
What does Parallel's Scale plan cost?
$1,000 a month for $1,000 in credits at standard API rates, or $10,000 a year for $12,000 in credits. Unused monthly credits expire at the end of each paid month. SpecStream has no plan to buy, and credit never expires.
Is there a cheaper alternative to the Parallel Search API?
Yes. SpecStream answers Parallel's own /v1/search call at $1 per 1,000 searches of up to 10 results in basic mode, a fifth of Parallel's $5, and 20 results cost $0.0022 against $0.011. The change is the base address and the key in x-api-key.
How much does Parallel FindAll cost?
$0.25 a run plus $0.03 a match on base, $2 plus $0.15 on core and $10 plus $1 on pro, with enrichments billed per match. SpecStream Deep Search charges 5¢, 15¢ or $1 a delivered row, with no fee per run.
Does SpecStream have an API and an MCP server?
Yes. The Search API is GET https://api.specstream.ai/v1/web-search at $5 per 1,000 searches, or $1 with speed=fast, and the MCP server is at https://api.specstream.ai/mcp over Streamable HTTP, with OAuth or an API key.

Sources

SpecStream wrote this page. Each Parallel price comes from Parallel's own pricing page and pricing docs, read on Nov 12, 2026.

  1. SpecStream pricingspecstream.ai/pricing, read Nov 12, 2026
  2. SpecStream for agentsspecstream.ai/agents.md, read Nov 12, 2026
  3. SpecStream benchmarksspecstream.ai/benchmarks, read Nov 12, 2026
  4. Parallel pricingparallel.ai/pricing, read Nov 12, 2026
  5. Parallel pricing docsdocs.parallel.ai/getting-started/pricing, read Nov 12, 2026
  6. Parallel search modesdocs.parallel.ai/search/modes, read Nov 12, 2026
  7. Parallel advanced search settingsdocs.parallel.ai/search/advanced-search-settings, read Nov 12, 2026
  8. Brave Search APIbrave.com/search/api/, read Nov 12, 2026
  9. Exa pricingexa.ai/docs/admin/pricing.md, read Nov 12, 2026
  10. Tavily creditsdocs.tavily.com/documentation/api-credits.md, read Nov 12, 2026
  11. You.com pricingyou.com/pricing, read Nov 12, 2026