NewYour Exa or Tako SDK, up to 5× cheaper.

Compared with Event Registry

vsEvent Registry

The Event Registry (NewsAPI.ai) alternative for AI agents

Event Registry is built for news analytics: concepts, events and an archive back to 2014. typesearch is built for agents that search the news now: a calibrated relevance probability on every result, the top ones read before they’re ranked, typed answers per article — from $1.40 per 1,000 searches, with no monthly plan.

Checked Sep 25, 2026 · A news API with events, concepts and a 2014+ archive

At the same volume

  • 5K · 5,000 requests a month

    $7/ month

    typesearch, fast mode

    Event Registry: $90−92%

    $18 per 1,000 · 5,000 tokens a month

  • 20K · 20,000 requests a month

    $28/ month

    typesearch, fast mode

    Event Registry: $220−87%

    $11 per 1,000 · 20,000 tokens a month

  • 100K · 100,000 requests a month

    $140/ month

    typesearch, fast mode

    Event Registry: $700−80%

    $7 per 1,000 · 100,000 tokens a month

  • 500K · 500,000 requests a month

    $700/ month

    typesearch, fast mode

    Event Registry: $3,000−76%

    $6 per 1,000 · 500,000 tokens a month

Event Registry counts tokens: a search of recent articles is 1 token and returns up to 100 articles; an archive search costs 5 tokens per year searched. Extra tokens are $0.015 and tokens don’t roll over. typesearch returns up to 50 judged results per search, with no plan.

Checked Sep 25, 2026 · Sources: newsapi.ai/plans

Side by side

What Event Registry publishes, next to what typesearch does. Each of their facts links to where we read it.

Side by sideEvent Registrytypesearch
PricingMonthly token plans: $90 to $3,000; a recent search is 1 token, an archive search 5 per year searched Pay as you go, per search: $1.40 fast, $2.20 normal, $5.60 deep, per 1,000. No plan, no minimum
Free plan2,000 tokens once, last 30 days only; testing only, no commercial use $5 in credit once, valid 30 days — not limited to development or testing
How far backSince 2014 on paid plans A rolling 30-day window: built for what’s happening now
Coverage150,000+ outlets, 50+ languages (their figures) Sources in 130+ countries and 30+ languages; counts by country and language in GET /v1/sources
Full article textYes, on every plan No: title, the outlet’s standfirst and short verbatim excerpts, always with the link
Per requestUp to 100 articles or 50 events Up to 50 judged results; up to five queries per request
RankingSort by date, relevance or source importance A calibrated relevance probability per result; the top ones read and verified in normal and deep
EnrichmentEntities and concepts, categories, event clusters, duplicates; sentiment in English Your own typed questions per result, tone, same-story grouping, who published first
Rate limitUp to 5 concurrent requests 10 requests/second; 50 with a monthly plan
SDKsPython and Node.js OpenAPI 3.1; works with the official Exa, Tako and Perplexity SDKs
MCP serverYes, local (npm) Yes, remote: api.typesearch.ai/mcp

Checked Sep 25, 2026 · Sources: newsapi.ai/plans, newsapi.ai/documentation, newsapi.ai/terms, newsapi.ai/documentation

How the parameters map

Each Event Registry parameter and what it becomes on typesearch. You rewrite one request; your agent’s logic stays.

Event Registry parameterStatusOn typesearch
keywordMappedquery. Or up to five queries, judged together.
conceptUri, categoryUriLimitedName the concept in the query; the model judges meaning, not just keywords.
sourceUriMappedinclude_domains, up to 20.
langLimitedQuery in the language you want, or restrict to regional outlets with include_domains.
dateStart, dateEndMappedpublished_after, published_before, within the last 30 days.
articlesCount, articlesPageLimitedmax_results, up to 50. No pagination.
articlesSortByLimitedRanked by a calibrated relevance probability; published_at to sort by date.
isDuplicateFilterMappeddedupe: true groups the same story, with the duplicates attached.
minSentiment, maxSentimentMappedtone: true — positive, neutral or negative toward your query, in any language.
getEventsMapped/v1/similar: every article about the same story as a URL.
articleBodyLenLimitedNever the full article: verbatim highlights in normal and deep.
apiKeyMappedAuthorization: Bearer <key>
articles.results[]: title, url, body, dateTimePub, source.titleMappedresults[]: title, url, snippet, published_at, source

What you gain

  • Pay as you go from $1.40 per 1,000 searches, against $18 per 1,000 recent searches on the 5K plan
  • No tokens that expire at the end of the month
  • A calibrated relevance probability on every result
  • Your own typed questions answered per article; tone in any language
  • A free credit you can use beyond testing

When Event Registry is the better fit

We’d rather you pick right the first time. Keep Event Registry for these:

  • You need an archive back to 2014, or searches across years.
  • You work with concepts and event clusters rather than keyword queries.
  • You need the full text of every article.

The migration

The whole migration

Before and after: the request, the key header and the fields you read. Everything else in your code stays.

import requests
 
-r = requests.post("https://eventregistry.org/api/v1/article/getArticles",
- json={"apiKey": ER_API_KEY, "keyword": "lithium royalties in Chile", "dateStart": "2026-09-18", "articlesCount": 20},
-).json()
+r = requests.post("https://api.typesearch.ai/v1/search",
+ headers={"Authorization": f"Bearer {TYPESEARCH_API_KEY}"},
+ json={"query": "lithium royalties in Chile", "days": 7, "max_results": 20, "mode": "fast"},
+).json()
 
-for a in r["articles"]["results"]:
- print(a["title"], a["url"], a["dateTimePub"], a["source"]["title"])
+for a in r["results"]:
+ print(a["title"], a["url"], a["published_at"], a["source"], a["score"])
beforeafter

Switching from Event Registry

How is typesearch different from Event Registry?

Event Registry annotates every article with concepts, categories and the event it belongs to, and keeps an archive since 2014. typesearch judges results for your query in real time — a calibrated probability, the top ones read — and answers your own typed questions per article.

How much does typesearch cost compared with Event Registry?

At the volume of Event Registry’s 5K plan (5,000 requests a month), typesearch costs $7 in fast mode, against $90. At the volume of Event Registry’s 20K plan (20,000 requests a month), typesearch costs $28 in fast mode, against $220. At the volume of Event Registry’s 100K plan (100,000 requests a month), typesearch costs $140 in fast mode, against $700. At the volume of Event Registry’s 500K plan (500,000 requests a month), typesearch costs $700 in fast mode, against $3,000. Results served from the cache are free, and there are no seats or minimums.

How do I switch from Event Registry?

Rewrite one request: the endpoint, the key header and a few parameter names change — see the table above. The fields you read map one to one: body → snippet and highlights, dateTimePub → published_at, source.title → source.

When should I stay with Event Registry?

If you need an archive back to 2014, concept and event search, full article text, or 100 articles per request. You can also use both: typesearch for what your agent searches now, Event Registry for the rest.

Switch from Event Registry

A key in a minute, $5 in free credit, one request to rewrite.