benchmarks/company news/newscatcher pricing
not on the current board · facts from newscatcherapi.com

NewsCatcher API pricing and entity search

NewsCatcher is a news search API priced in credits, with a usable free tier and an entity filter that is sold on the Enterprise plan. It is not on the current company-news board. This page covers the credits, the limits, how the ORG_entity_name filter works, and what the benchmark says.

The existing NewsCatcher alternatives page ranks the measured search APIs. This one is about NewsCatcher itself.

Benchmark last measured . 300 questions.

Credits, plans and limits

  • Free. 10,000 credits every 30 days, spent first, no rollover. Last 30 days of news. One call a second.
  • Pay as you go. Top up from $25 for 25,000 credits, valid 12 months. Up to 10 calls a second.
  • Enterprise. Custom. News back to 2019, entities, sentiment, translations, custom rate limits.
  • What a call costs. News search at low, medium or high detail: 1, 5 or 10 credits. Everything endpoint: 5 credits a page. Only successful work is charged.

At the pay-as-you-go rate a one-credit search is $0.001, or $1 per 1,000 searches, which is cheaper than any paid web search API on the board. The catch is the detail level: a one-credit call returns less per article than a ten-credit one, and the entity filter is not on these plans at all.

Vendor prices, limits and parameter names were read from the linked pages on 18 November 2026. They are the vendor's list terms, not measurements.

How ORG_entity_name works

Search is GET /api/search on the v3 host. The plain q parameter matches text. The entity parameters match what NewsCatcher's NLP found: ORG_entity_name for organizations, PER_entity_name, LOC_entity_name and MISC_entity_name for the rest. Each takes AND, OR, NOT and quoted exact matches. A query of "Apple Inc" OR Microsoft returns articles where the tagger identified either company as an organization, which is different from articles that contain the word apple.

Set include_nlp_data to true to get the nlp object on each article, with ner_ORG, ner_PER, ner_LOC and ner_MISC as lists of entity names and mention counts. has_nlp drops articles without it. Two limits: NLP enrichment only exists for articles indexed from July 2023 onward, and it has to be enabled in your plan, which on the pricing page means Enterprise. page_size goes to 1,000 and clustering_enabled groups near-duplicate articles.

There is no company identifier. The filter is the organization string the tagger produced, so two spellings of one company are two values unless you OR them. That is one step short of a concept URI at Event Registry or a graph node at Diffbot, and a long way past a keyword at NewsAPI.org.

What the benchmark says

NewsCatcher is not on the current board. The runner has a NewsCatcher arm configured, POST /api/search with page_size=10 and the question as plain text, and the free tier covers the 300 questions, so it is part of the re-run that puts every vendor in one window. Below is the top of the web search band on the same questions.

Top five measured web search configurations on company-news questions

RankProviderAccuracyAnswer recall @5Mean search time$ / 1k queries$ / 1k correct
1Exa (type=fast)POST /search · type=fast99.3%99.3%652ms$7.00$7.05
3Exa (type=instant)POST /search · type=instant97.7%97.3%398ms$7.00$7.17
4PerplexityPOST /search · search_context_size=low97.3%98.3%1.38s$5.00$5.14
5Linkup (depth=fast)POST /v1/search · depth=fast · outputType=searchResults96.7%94.7%1.57s$5.00$5.17

Exa (type=fast) answers 99.3% of the questions from text alone. That is the lookup bar. NewsCatcher's case is the stream: the same company every day, tagged, clustered and filtered by organization, at a credit price below the search APIs.

Questions answered on this page

How much does NewsCatcher cost?

Every account gets 10,000 free credits every 30 days; they are spent first and do not roll over. Pay as you go starts at $25 for 25,000 credits, valid for 12 months. A news search costs 1, 5 or 10 credits a call depending on the detail level, and an Everything page costs 5. Enterprise is custom pricing and is where entities, sentiment, translations and history back to 2019 live. Prices are in US dollars.

What is in the free tier?

The last 30 days of news, one call a second, and the 10,000 monthly credits. At one credit a low-detail search, that is 10,000 searches a month. The free tier does not include NLP enrichment, so the entity filter is not available on it.

How does the entity filter work?

NewsCatcher runs named-entity recognition over each article and stores the organizations, people, locations and other entities it found, with mention counts. The v3 search endpoint takes ORG_entity_name, PER_entity_name, LOC_entity_name and MISC_entity_name, each accepting AND, OR, NOT and quoted exact matches, so "Apple Inc" OR Microsoft returns articles where the NLP tagged either organization. Set include_nlp_data to get the nlp object back; has_nlp drops articles without one. Articles indexed before July 2023 have no NLP data.

Has NewsCatcher been benchmarked?

Not on the current board. The benchmark runner has a NewsCatcher search arm configured, POST /api/search with page_size=10, and the free tier covers the 300 questions, so it is part of the re-run that puts every vendor in one window.

NewsCatcher or a web search API for company news?

For one factual question, the measured web search APIs are the evidence: Exa (type=fast) answered 99.3% from plain text. For a monitoring feed on a company whose name is ambiguous or spelled several ways, the ORG_entity_name filter is the feature the search APIs do not have, and it needs an Enterprise plan.