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

Compared with Perigon

vsPerigon

The Perigon alternative for AI agents

Perigon bundles news with entities, story clusters and summaries — on startup licenses, then commercial plans from $24,000 a year. typesearch is pay as you go from $1.40 per 1,000 searches, with a calibrated relevance probability on every result and typed answers in the same call.

Checked Sep 25, 2026 · A news context platform with rich enrichment

At the same volume

  • Basic · 10,000 requests a month

    $14/ month

    typesearch, fast mode

    Perigon: $250−94%

    $25 per 1,000 · Startup license, up to 1 year; 1-hour delay

  • Plus · 50,000 requests a month

    $70/ month

    typesearch, fast mode

    Perigon: $550−87%

    $11 per 1,000 · Startup license, up to 1 year

Perigon’s Basic and Plus prices are for eligible startups, for up to a year, for prototyping or products not yet in market; after that, commercial plans start at $24,000 a year. One request returns up to 50 (Basic) or 100 (Plus) articles; up to 50 judged results on typesearch.

Checked Sep 25, 2026 · Sources: perigon.io/products/pricing/apis

Side by side

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

Side by sidePerigontypesearch
PricingBasic $250 and Plus $550 for eligible startups, up to a year; commercial plans from $24,000 a year Pay as you go, per search: $1.40 fast, $2.20 normal, $5.60 deep, per 1,000. No plan, no minimum
Free plan150 requests a month; personal, non-commercial use $5 in credit once, valid 30 days — not limited to development or testing
How far back3 months free, 6 months Basic, 3 years Plus, 10+ years commercial A rolling 30-day window: built for what’s happening now
Coverage230,000+ sources (their figure; languages vary across their pages) Sources in 130+ countries and 30+ languages; counts by country and language in GET /v1/sources
Full article textDepends on your use case and license No: title, the outlet’s standfirst and short verbatim excerpts, always with the link
Per request25, 50 or 100 articles by plan Up to 50 judged results; up to five queries per request
RankingSort by relevance or date A calibrated relevance probability per result; the top ones read and verified in normal and deep
EnrichmentEntities, sentiment, AI summaries, story clustering, vector search Your own typed questions per result, tone, same-story grouping, who published first
Rate limit1, 2 or 4 requests/second by plan 10 requests/second; 50 with a monthly plan
SDKsGo, TypeScript and Python OpenAPI 3.1; works with the official Exa, Tako and Perplexity SDKs
MCP serverYes, remote, on every plan Yes, remote: api.typesearch.ai/mcp

Checked Sep 25, 2026 · Sources: perigon.io/products/pricing/apis, perigon.io, perigon.io/openapi.json, perigon.io/docs/api/perigon-sdks

How the parameters map

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

Perigon parameterStatusOn typesearch
qSamequery. Or up to five queries, judged together.
title, desc, contentLimitedWe judge title and standfirst, and read the top articles in normal and deep.
from, toMappedpublished_after, published_before, within the last 30 days.
source, excludeSourceMappedinclude_domains, exclude_domains, up to 20 each.
language, countryLimitedQuery in the language you want, or restrict to regional outlets with include_domains.
category, topicMappedsections, or name the topic in the query.
showReprintsMappeddedupe: true groups reprints of the same story, with the duplicates attached.
positiveSentimentFrom…Mappedtone: true — positive, neutral or negative toward your query.
sortBy, size, pageLimitedRanked by relevance; max_results up to 50; no pagination.
x-api-keyMappedAuthorization: Bearer <key>
articles[]: title, url, description, pubDate, source.domainMappedresults[]: title, url, snippet, published_at, source

What you gain

  • Commercial use from day one, pay as you go — no startup license to outgrow
  • From $1.40 per 1,000 searches, against $25 per 1,000 on Basic
  • Real time on every request (Basic runs an hour behind)
  • Your own typed questions answered per article, on every request
  • Works with the Exa, Tako and Perplexity SDKs

When Perigon is the better fit

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

  • You need entity, company and journalist data linked to every article.
  • You want story clusters and AI summaries from the API itself.
  • You need an archive of several years.

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.get("https://api.perigon.io/v1/articles/all",
- headers={"x-api-key": PERIGON_API_KEY},
- params={"q": "lithium royalties in Chile", "from": "2026-09-18", "size": 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"]:
- print(a["title"], a["url"], a["pubDate"], a["source"]["domain"])
+for a in r["results"]:
+ print(a["title"], a["url"], a["published_at"], a["source"], a["score"])
beforeafter

Switching from Perigon

How is typesearch different from Perigon?

Perigon enriches news with entities, story clusters, summaries and vector search, under licenses that depend on the stage of your company. typesearch is one commercial price list for everyone: judged results with a calibrated probability, the top ones read, and your own typed questions per article.

How much does typesearch cost compared with Perigon?

At the volume of Perigon’s Basic plan (10,000 requests a month), typesearch costs $14 in fast mode, against $250. At the volume of Perigon’s Plus plan (50,000 requests a month), typesearch costs $70 in fast mode, against $550. Results served from the cache are free, and there are no seats or minimums.

How do I switch from Perigon?

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: description → snippet, pubDate → published_at, source.domain → source.

When should I stay with Perigon?

If you need entity, company and journalist data, story clusters and summaries, or an archive of 10+ years. You can also use both: typesearch for what your agent searches now, Perigon for the rest.

Switch from Perigon

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