Web Search / Best search API for company research
Independent comparison · 2026

Best Search API for Company Research (2026): Search and Scrape F1, Speed & Cost

Best search API for company research by F1: Exa deep, 48.2% at $12.00/1,000 queries. Fastest: SERP (RapidAPI), 3.50s. Cheapest: TinyFish, $0.00/1,000 queries within free-tier limits.

Compare 12 providers, including Exa, Tavily, Brave, Parallel and Firecrawl, across 20 configurations on 45 questions. Compare search APIs for company research, with search and page scrape, scored by F1 on company questions with more than one constraint.

Compare all 20 configurations → · Open benchmark code ↗ · Public dataset sample ↗

Updated 23 September 2026. Last measured 14 September 2026.

Best search API for company research

Mean F1 on 45 company-discovery questions with multiple constraints. The same agent can search more than once and scrape the page. F1 balances precision and recall. Search latency and total task latency are shown separately. Prices are the benchmark-recorded rate basis, with the same configurations used for the quality results.

45 questions · 20 configurations · ordered by F1
RankProvider & configurationF1Mean search timeRecorded $ / 1,000 queriesMedian task timeMedian task costOfficial docs
1Exa deepexa-deepPOST /search type=deep48.2%47.08s$12.00type=deepScrape $0.0010 per page.95.80s$0.68Official docs ↗
2Perplexitysearch_context_size=highPOST /search46.6%13.98s$5.00Search API · POST /searchScrape included in the measured task cost when the provider billed it.53.89s$0.50Official docs ↗
3Exa instantexa-instantPOST /search type=instant44.9%5.90s$7.00type=instant · up to 10 resultsScrape $0.0010 per page.52.51s$0.65Official docs ↗
4Parallel basicparallel-basicPOST /v1/search mode=basic42.3%21.58s$5.00mode=basic · 10 resultsScrape $0.0010 per page.68.62s$1.09Official docs ↗
5Parallel advancedparallel-advancedPOST /v1/search mode=advanced42.2%36.66s$5.00mode=advanced · 10 resultsScrape $0.0010 per page.80.87s$0.60Official docs ↗
6Linkup standardlinkup-standardPOST /v1/search depth=standard42.0%27.77s$5.00depth=standard · searchResultsScrape $0.0010 per page.81.01s$0.91Official docs ↗
7Tavily advancedtavily-advancedPOST /search search_depth=advanced41.0%44.65s$16.002 credits · $0.008 PAYGScrape $0.0032 per page.92.63s$0.90Official docs ↗
8Linkup fastlinkup-fastPOST /v1/search depth=fast39.9%17.28s$5.00depth=fast · searchResultsScrape $0.0010 per page.68.02s$0.90Official docs ↗
9Parallel fastparallel-fastPOST /v1/search mode=fast39.3%12.27s$1.00mode=fast · 10 resultsScrape $0.0010 per page.55.08s$0.44Official docs ↗
10Youextraction_mode=highlightsPOST /v1/search38.5%7.55s$5.00highlights included in Web SearchScrape $0.0010 per page.45.63s$0.86Official docs ↗
11Youextraction_mode=highlights · knowledge=corePOST /v1/search38.5%9.18s$5.00highlights · knowledge=core · same Web Search price assumptionScrape $0.0010 per page.48.52s$0.89Official docs ↗
12Tavily basictavily-basicPOST /search search_depth=basic38.1%20.30s$8.001 credit · $0.008 PAYGScrape $0.0032 per page.61.97s$0.61Official docs ↗
13Parallel turboparallel-turboPOST /v1/search mode=turbo36.0%5.38s$1.00mode=turbo · 10 resultsScrape $0.0010 per page.48.07s$0.41Official docs ↗
14FirecrawlfirecrawlPOST /v2/search33.2%40.23s$5.002 credits / 10 resultsScrape $0.0025 per page.82.06s$0.29Official docs ↗
15Nimblesearch_depth=standard · full_content=false · focus=general · extract formats=[markdown]POST /v2/search31.3%23.00s$5.00search_depth=standard · full_content=falseScrape included in the measured task cost when the provider billed it.59.23sOfficial docs ↗
16TinyFishtinyfishGET api.search.tinyfish.ai30.2%26.59s$0.0030 req/min capScrape $0.0000 per page.63.37s$0.22Official docs ↗
17Brave SearchbraveGET /res/v1/web/search29.4%7.15s$5.00Recorded search-request rate; see official terms.Scrape included in the measured task cost when the provider billed it.45.13s$0.28Official docs ↗
18Nimblesearch_depth=lite · full_content=false · focus=general · extract formats=[markdown]POST /v2/search25.8%53.50s$1.10search_depth=lite · full_content=falseScrape included in the measured task cost when the provider billed it.72.36sOfficial docs ↗
19Seltzseltz-companiesPOST /v1/search scope=companies16.3%9.30s$5.00Recorded search-request rate; see official terms.Scrape included in the measured task cost when the provider billed it.60.10s$1.74Official docs ↗
20SERP (RapidAPI)serpGET google-search74.p.rapidapi.com0.0%3.50s$3.00Pro overage $0.003/requestScrape included in the measured task cost when the provider billed it.33.44s$0.10Official docs ↗

Choose the best search API for company research

Company research in this benchmark

Questions ask for a set of companies matching more than one constraint, such as investor, geography, accelerator or funding stage. The agent can search and scrape pages. F1 is the rank. That is the measured evidence for a search API for company research on this page.

Precision and recall for a company set

A company-research answer is a set, so an extra wrong company lowers precision and a missed company lowers recall. F1 falls when either side is weak. Use the table to see which search and scrape configuration recovers the set with fewer misses and fewer extras.

What the cost columns include

Recorded search price is the search call per 1,000 queries. Scraping the company page can add a second charge. Median task cost includes model tokens and the vendor spend measured on the run. A cheap search call can still be an expensive research task if the agent searches many times.

Check coverage on your own company questions

This set is held-out company discovery, public in sample form, with scoring questions kept aside. Rebuild a short list from your market, run the same agent against two configurations, and compare the companies and the URLs. Official docs in the table document the request that was measured.

Benchmark method and source data

Mean F1 on 45 company-discovery questions with multiple constraints. The same agent can search more than once and scrape the page. F1 balances precision and recall. Search latency and total task latency are shown separately. Each row is one measured configuration. Missing measurements appear as a dash.

Open benchmark code ↗ · Public dataset sample ↗ · All benchmark methods →

Compare search and search-and-scrape results →

Common questions

What is the best search API for company research?

Best search API for company research by F1: Exa deep, 48.2% at $12.00/1,000 queries. Fastest: SERP (RapidAPI), 3.50s. Cheapest: TinyFish, $0.00/1,000 queries within free-tier limits.

Which benchmark does the main ranking use?

Mean F1 on 45 company-discovery questions with multiple constraints. The same agent can search more than once and scrape the page. F1 balances precision and recall. Search latency and total task latency are shown separately.

What does the price per 1,000 queries include?

The recorded search endpoint’s request price. Model-token spend, page scrape and plan conditions are separate. Rows with a task measurement show median task cost in its own column. Check the linked provider terms for current pricing.

Related search API comparisons