benchmarks/company-news/general web search APIs
300 questions · 19 endpoints · general web search

Web search APIs for company news

Query in, ranked web results out. These endpoints search the open web rather than a curated index, and are ranked here against each other with the news indexes held out, so the comparison is between substitutes.

Measured on the same questions, the strongest dedicated news index, Seltz (scope=news), reaches 58.3%. Every endpoint below beats it, which is why the two are ranked apart.

News index APIs for company news · Most accurate · Fastest · Cheapest · Full company news board · Web search API benchmark

Open CodeHarness, scoring and vendor runners in openbenchmarks-labs/company-news.github →public dataset100 public company-news questions and ground truth answers on Hugging Face as openbenchmarks/OB-News-Websearch.huggingface →

Web search APIs for company news, all metrics

19 endpoints, ranked by $ / 1k correct on 300 questions.

Web search APIs for company news, ranked by $ / 1k correct on the shared question set
RankVendorEndpoint & configuration$ / 1k correctAccuracyAR@1AR@5LatencySnippet tokensTotal $Official docsCost per 1,000 queries
1TinyFishweb searchGET api.search.tinyfish.aifree · 30 req/min capFree30 req/min cap92.0%74.3%90.7%2.62s441FreeOfficial docs$0free · 30 req/min, 500/hour
2Parallelweb searchPOST /v1/searchmode=fast$1.1686.0%44.3%79.0%942ms1,839$0.30Official docs$1 / 1kmode=fast · 10 results
3Parallelweb searchPOST /v1/searchmode=turbo$1.4071.3%45.3%66.0%348ms1,853$0.30Official docs$1 / 1kmode=turbo · 10 results
4Nimbleweb searchPOST /v2/searchsearch_depth=lite · full_content=false · focus=general$1.4675.3%64.0%75.3%3.15s413$0.33Official docs$1.1 / 1ksearch_depth=lite · full_content=false
5SERPgoogle searchGET google-search74.p.rapidapi.comlimit=10$3.1396.0%78.0%95.0%751ms497$0.90Official docs$3 / 1kPro overage $0.003/request
6Perplexityweb searchPOST /searchsearch_context_size=low$5.1497.3%91.7%98.3%1.38s476$1.50Official docs$5 / 1kSearch API · POST /search
7Linkupweb searchPOST /v1/searchdepth=fast · outputType=searchResults$5.1796.7%79.0%94.7%1.57s3,022$1.50Official docs$5 / 1kdepth=fast · searchResults
8Firecrawlweb searchPOST /v2/search$5.2495.3%77.7%96.7%510ms678$1.50Official docs$5 / 1k2 credits / 10 results
9Brave Searchweb searchPOST /res/v1/llm/contextcount=10$5.3294.0%81.0%94.7%601ms2,064$1.50Official docs$5 / 1kSearch plan · LLM Context
10Parallelweb searchPOST /v1/searchmode=basic$5.3693.3%55.0%92.0%1.68s2,330$1.50Official docs$5 / 1kmode=basic · 10 results
11Brave Searchweb searchGET /res/v1/web/searchcount=10 · result_filter=web$5.3693.3%79.3%91.7%630ms817$1.50Official docs$5 / 1k
12Nimbleweb searchPOST /v2/searchsearch_depth=standard · full_content=false · focus=general$5.3893.0%86.3%94.3%861ms2,730$1.50Official docs$5 / 1ksearch_depth=standard · full_content=false
13Linkupweb searchPOST /v1/searchdepth=standard · outputType=searchResults$5.4392.0%67.3%90.3%2.55s2,983$1.50Official docs$5 / 1kdepth=standard · searchResults
14Youweb searchPOST /v1/searchextraction_mode=highlights · knowledge=core$5.4392.0%67.3%89.7%889ms2,862$1.50Official docs$5 / 1khighlights · knowledge=core · same Web Search price assumption
15Youweb searchPOST /v1/searchextraction_mode=highlights$5.5190.7%72.0%89.7%628ms2,837$1.50Official docs$5 / 1khighlights included in Web Search
16Exaweb searchPOST /searchtype=fast$7.0599.3%95.0%99.3%652ms1,987$2.10Official docs$7 / 1ktype=fast · up to 10 results
17Exaweb searchPOST /searchtype=instant$7.1797.7%80.0%97.3%398ms2,128$2.10Official docs$7 / 1ktype=instant · up to 10 results
18Tavilyweb searchPOST /searchsearch_depth=basic$9.1387.7%82.0%87.7%1.88s1,639$2.40Official docs$8 / 1k1 credit · $0.008 PAYG
19Tavilyweb searchPOST /searchsearch_depth=advanced$17.2093.0%65.0%92.7%4.29s2,210$4.80Official docs$16 / 1k2 credits · $0.008 PAYG

$ / 1k correct is $ / 1k queries divided by extracted-answer accuracy. Ranked by $ / 1k correct. Cost per 1,000 queries is the published PAYG list price, linked to the vendor pricing page. Not promotional packs or volume discounts. TinyFish Search is free with a 30 requests/min cap — $0 is not unlimited throughput.

News index APIs for company news, same questions

Measured on the identical question set with the same judge, ranked separately because it is a different mechanism.

News index APIs for company news, for comparison
RankVendorEndpoint & configuration$ / 1k correctAccuracyAR@1AR@5LatencySnippet tokensTotal $Official docsCost per 1,000 queries
1Seltznews indexPOST /v1/searchscope=news$8.5758.3%44.3%60.0%403ms2,919$1.50Official docs$5 / 1k
2Autoboundcompany news eventsPOST /v1/companies/enrichdomain · signal_types=news · signal_subtypes · limit=10$120.7630.0%21.0%25.0%146ms879$10.87Official docs$9.5 / 1k creditsStarter $19 / 2,000 · 2 credits/signal
3PredictLeadscompany news eventsGET /api/v3/companies/{domain}/news_eventsLLM category pick · categories[] · limit=10$193.5220.7%14.0%20.0%398ms484$12.00Official docs$40 / 1k$0.04 / credit · 1 credit/request
4Datahyenacompany news eventsGET /v1/companies/timelinedomain · include=funding|acquisitions|exec_moves$716.075.3%5.3%6.7%187ms22$11.45Official docs$50 / 1k credits$25 / 500 credits · 1 credit/record

$ / 1k correct is $ / 1k queries divided by extracted-answer accuracy. Ranked by $ / 1k correct. Cost per 1,000 queries is the published PAYG list price, linked to the vendor pricing page. Not promotional packs or volume discounts. TinyFish Search is free with a 30 requests/min cap — $0 is not unlimited throughput.

Open the news index apis for company news page →

Web search APIs for company news

What is the best web search API for company news?

Exa (type=fast), at 99.3% correct on 300 company-news lookups — the highest of any endpoint measured, on either mechanism. Exa (type=instant) follows at 97.7%. The strongest dedicated news index reaches 58.3% on the identical questions, so a general web search API is the more accurate choice for this task.

What counts as a general web search API here?

An endpoint that takes a query and searches the open web, returning ranked results the caller reads. Curated news indexes are measured on the same questions but ranked separately, because they are a different mechanism.

How are these endpoints made comparable?

Every endpoint receives the identical user query, may return at most 10 results, and is scored by one judge against ground truth pinned to official sources. Nothing is rewritten per vendor. 300 questions, one run.

Why do some vendors appear more than once?

Several expose more than one search depth or mode, and those behave differently enough on speed and cost to be worth separating. Each endpoint is its own row, with its request configuration shown beside it.